To see the other types of publications on this topic, follow the link: Federated network.
Dissertations / Theses on the topic 'Federated network'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 44 dissertations / theses for your research on the topic 'Federated network.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Kulkarni, Shweta Samir. "SECURE MIDDLEWARE FOR FEDERATED NETWORK PERFORMANCE MONITORING." The Ohio State University, 2013. http://rave.ohiolink.edu/etdc/view?acc_num=osu1366333088.
Full text
2
Maka, Stephan. "Design and Implementation of a Federated Social Network." Master's thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2011. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-75477.
Full text
3
Felt, Aaron J. "Federated ground station network model and interface specification." Thesis, Monterey, California: Naval Postgraduate School, 2014. http://hdl.handle.net/10945/44558.
Full textAbstract:
Approved for public release; distribution is unlimitedThis thesis solves the problem of a lack of a complete, simple ground station network interface standard. A federated satellite ground station network (FGN) model and computer interface are developed that extend the use of ground stations to external users across the Internet. This should allow for reuse of existing ground stations, reducing costs and complexity of space missions. An improved model describing FGNs is proposed that defines a hierarchy of the components of the network, allowing for scalability and unified interfaces, and simplifying the process of using FGN resources. This model, which we call the Improved FGN model, is used to develop security schemes that are simple but effective. Simple but effective security schemes are then developed for this Improved FGN model, along with a standardized SOFtware interface. This interface connects external users to the network in order to extend ground station hardware to remote users as well as to simplify scheduling for the resource owners in a network. Different middleware frameworks are compared, and Apache Thrift is selected as the best fit for an FGN. This interface is then described and demonstrated with a reference implementation in Python. Recommendations for future improvements of this interface standard are discussed.
4
Demirci, Turan. "Federated Simulation Of Network Performance Using Packet Flow Modeling." Phd thesis, METU, 2010. http://etd.lib.metu.edu.tr/upload/2/12611704/index.pdf.
Full textAbstract:
Federated approach for the distributed simulation of a network, is an alternative method that aims to combine existing simulation models and software together using a Run Time Infrastructure (RTI), rather than building the whole simulation from scratch. In this study, an approach that significantly reduces the inter-federate communication load in federated simulation of communication networks is proposed. Rather than communicating packet-level information among federates, characteristics of packet flows in individual federates are dynamically identified and communicated. Flow characterization is done with the Gaussian Mixtures Algorithm (GMA) using a Self Organizing Mixture Network (SOMN) technique. In simulations of a network partitioned into eight federates in space parallel manner, it is shown that significant speedups are achieved with the proposed approach without unduly compromising accuracy.
5
Cetin, Burak. "Wireless Network Intrusion Detection and Analysis using Federated Learning." Youngstown State University / OhioLINK, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=ysu1588778320687729.
Full text
6
Xu, Ran. "Federated Sensor Network architectural design for the Internet of Things (IoT)." Thesis, Loughborough University, 2013. https://dspace.lboro.ac.uk/2134/13453.
Full textAbstract:
An information technology that can combine the physical world and virtual world is desired. The Internet of Things (IoT) is a concept system that uses Radio Frequency Identification (RFID), WSN and barcode scanners to sense and to detect physical objects and events. This information is shared with people on the Internet. With the announcement of the Smarter Planet concept by IBM, the problem of how to share this data was raised. However, the original design of WSN aims to provide environment monitoring and control within a small scale local network. It cannot meet the demands of the IoT because there is a lack of multi-connection functionality with other WSNs and upper level applications. As various standards of WSNs provide information for different purposes, a hybrid system that gives a complete answer by combining all of them could be promising for future IoT applications. This thesis is on the subject of `Federated Sensor Network' design and architectural development for the Internet of Things. A Federated Sensor Network (FSN) is a system that integrates WSNs and the Internet. Currently, methods of integrating WSNs and the Internet can follow one of three main directions: a Front-End Proxy solution, a Gateway solution or a TCP/IP Overlay solution. Architectures based on the ideas from all three directions are presented in this thesis; this forms a comprehensive body of research on possible Federated Sensor Network architecture designs. In addition, a fully compatible technology for the sensor network application, namely the Sensor Model Language (SensorML), has been reviewed and embedded into our FSN systems. The IoT as a new concept is also comprehensively described and the major technical issues discussed. Finally, a case study of the IoT in logistic management for emergency response is given. Proposed FSN architectures based on the Gateway solution are demonstrated through hardware implementation and lab tests. A demonstration of the 6LoWPAN enabled federated sensor network based on the TCP/IP Overlay solution presents a good result for the iNET localization and tracking project. All the tests of the designs have verified feasibility and achieve the target of the IoT concept.
7
Lu, Zonghao. "A case study about different network architectures in Federated Machine Learning." Thesis, Uppsala universitet, Institutionen för informationsteknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-425193.
Full textAbstract:
Modern artificial intelligence (AI) technology is developing rapidly in recent years. Data is an important factor driving the development of AI.With the development of mobile Internet, more and more data is generated in different fields every day along with data sensitivity issues. As asignificant part of personal privacy, personal data must be respected and protected. Federated learning (FL) is a machine learning technology that can protect privacy because it keeps everyone’s data local. Many types of research have already confirmed that the bottleneck of federated learning is the communication between the client and servers. Different ways of communication methods have various characteristics, resulting in differences inefficiency. We present a benchmark for our FL system using HTTP and gRPC communication protocol respectively to show that gRPC framework is faster and has better scalability than HTTP protocol mainly because of the different architectures and way of compacting message of these two different communication protocols. In addition, we found that the system may get crashed when the loads increased. A registration mechanism is proposed to deal with the problem of insufficient computing resources and apply a new model updatestrategy to make the training process finish in a shorter time.Tryckt av:
8
Sani, Lorenzo. "Unsupervised clustering of MDS data using federated learning." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2022. http://amslaurea.unibo.it/25591/.
Full textAbstract:
In this master thesis we developed a model for unsupervised clustering on a data set of biomedical data. This data has been collected by GenoMed4All consortium from patients affected by Myelodysplastic Syndrome (MDS), that is an haematological disease. The main focus is put on the genetic mutations collected that are used as features of the patients in order to cluster them. Clustering approaches have been used in several studies concerning haematological diseases such MDS. A neural network-based model was used to solve the task. The results of the clustering have been compared with labels from a "gold standard'' technique, i.e. hierarchical Dirichlet processes (HDP). Our model was designed to be also implemented in the context of federated learning (FL). This innovative technique is able to achieve machine learning objective without the necessity of collecting all the data in one single center, allowing strict privacy policies to be respected. Federated learning was used because of its properties, and because of the sensitivity of data. Several recent studies regarding clinical problems addressed with machine learning endorse the development of federated learning settings in such context, because its privacy preserving properties could represent a cornerstone for applying machine learning techniques to medical data. In this work will be then discussed the clustering performance of the model, and also its generative capabilities.
9
Gopalakrishnan, Aravind. "Network and Middleware Security for Enterprise Network Monitoring." The Ohio State University, 2012. http://rave.ohiolink.edu/etdc/view?acc_num=osu1339742304.
Full text
10
Vikström, Johan. "Comparing decentralized learning to Federated Learning when training Deep Neural Networks under churn." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-300391.
Full textAbstract:
Decentralized Machine Learning could address some problematic facets with Federated Learning. There is no central server acting as an arbiter of whom or what may benefit from Machine Learning models created by the vast amount of data becoming available in recent years. It could also increase the reliability and scalability of Machine Learning systems thereby drawing the benefit of having more data accessible. Gossip Learning is such a protocol, but has primarily been designed with linear models in mind. How does Gossip Learning perform when training Deep Neural Networks? Could it be a viable alternative to Federated Learning? In this thesis, we implement Gossip Learning using two different model merging strategies. We also design and implement two extensions to this protocol with the goal of achieving higher performance when training under churn. The training methods are compared on two tasks: image classification on the Federated Extended MNIST dataset and time- series forecasting on the NN5 dataset. Additionally, we also run an experiment where learners churn, alternating between being available and unavailable. We find that Gossip Learning performs slightly better in settings where learners do not churn but is vastly outperformed in the setting where they do.Decentraliserad Maskinginlärning kan lösa några problematiska aspekter med Federated Learning. Det finns ingen central server som agerar som domare för vilka som får gagna av Maskininlärningsmodellerna skapad av den stora mäng data som blivit tillgänglig på senare år. Det skulle också kunna öka pålitligheten och skalbarheten av Maskininlärningssystem och därav dra nytta av att mer data är tillgänglig. Gossip Learning är ett sånt protokoll, men det är primärt designat med linjära modeller i åtanke. Hur presterar Gossip Learning när man tränar Djupa Neurala Nätverk? Kan det vara ett möjligt alternativ till Federated Learning? I det här exjobbet implementerar vi Gossip Learning med två olika modelsammanslagningstekniker. Vi designar och implementerar även två tillägg till protokollet med målet att uppnå bättre prestanda när man tränar i system där noder går ner och kommer up. Träningsmetoderna jämförs på två uppgifter: bildklassificering på Federated Extended MNIST datauppsättningen och tidsserieprognostisering på NN5 datauppsättningen. Dessutom har vi även experiment då noder alternerar mellan att vara tillgängliga och otillgängliga. Vi finner att Gossip Learning presterar marginellt bättre i miljöer då noder alltid är tillgängliga men är kraftigt överträffade i miljöer då noder alternerar mellan att vara tillgängliga och otillgängliga.
11
Eriksson, Henrik. "Federated Learning in Large Scale Networks : Exploring Hierarchical Federated Learning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-292744.
Full textAbstract:
Federated learning faces a challenge when dealing with highly heterogeneous data and it can sometimes be inadequate to adopt an approach where a single model is trained for usage at all nodes in the network. Different approaches have been investigated to succumb this issue such as adapting the trained model to each node and clustering the nodes in the network and train a different model for each cluster where the data is less heterogeneous. In this work we study the possibilities to improve the local model performance utilizing the hierarchical setup that comes with clustering the participating clients in the network. Experiments are carried out featuring a Long Short-Term Memory network to perform time series forecasting to evaluate different approaches utilizing the hierarchical setup and comparing them to standard federated learning approaches. The experiments are done using a dataset collected by Ericsson AB consisting of handovers recorded at base stations in an European city. The hierarchical approaches didn’t show any benefit over common two-level approaches.Federated Learning står inför en utmaning när det gäller att hantera data med en hög grad av heterogenitet och det kan i vissa fall vara olämpligt att använda sig av en approach där en och samma modell är tränad för att användas av alla noder i nätverket. Olika approacher för att hantera detta problem har undersökts som att anpassa den tränade modellen till varje nod och att klustra noderna i nätverket och träna en egen modell för varje kluster inom vilket datan är mindre heterogen. I detta arbete studeras möjligheterna att förbättra prestandan hos de lokala modellerna genom att dra nytta av den hierarkiska anordning som uppstår när de deltagande noderna i nätverket grupperas i kluster. Experiment är utförda med ett Long Short-Term Memory-nätverk för att utföra tidsserieprognoser för att utvärdera olika approacher som drar nytta av den hierarkiska anordningen och jämför dem med vanliga federated learning-approacher. Experimenten är utförda med ett dataset insamlat av Ericsson AB. Det består av "handoversfrån basstationer i en europeisk stad. De hierarkiska approacherna visade inga fördelar jämfört med de vanliga två-nivåapproacherna.
12
Martins, Dalton Lopes. "Análise de redes sociais de colaboração científica no ambiente de uma federação de bibliotecas digitais." Universidade de São Paulo, 2012. http://www.teses.usp.br/teses/disponiveis/27/27151/tde-16042013-144121/.
Full textAbstract:
A produção científica de uma área do conhecimento aparece em diferentes formatos e é disponibilizada de forma essencialmente distribuída por entre revistas, anais, teses, dissertações e outros formatos característicos utilizados pela comunidade científica para a sistematização de seu discurso. Uma federação de bibliotecas digitais oferece uma arquitetura da informação que tem por finalidade facilitar a agregação de diferentes tipos de documentos disponibilizados, facilitando termos acesso a esses documentos, bem como a seus metadados descritores, formando, desse modo, verdadeiras estruturas de apoio ao desenvolvimento de pesquisas e análises científicas dos documentos que por ali circulam. Já a análise de redes sociais vem se mostrando um importante objeto de pesquisa da área da Ciência da Informação nas últimas décadas, tendo sido apropriada ainda de forma preliminar pela comunidade científica brasileira. Como forma de ampliar o conhecimento e experimentações com o uso da análise de redes sociais e identificar seu potencial analítico em relação ao que poderíamos coletar de informações de uma federação de bibliotecas digitais, tivemos por objetivo neste trabalho utilizar a análise de rede para mapear os padrões, tendências e estratégias de conectividade de dois planos de relacionamento entre pesquisadores: a coautoria em documentos oriundos de revistas científicas e a participação em bancas de defesas de teses e dissertações. Além disso, buscamos mapear as causas sociais e políticas dos padrões de rede identificados, colocando em evidência um uso crítico e contextualizado dos indicadores estruturais e dinâmicos de redes utilizados neste trabalho. Utilizamos como caso a biblioteca digital federada Univerciencia.org, uma biblioteca especializada na área de Ciências da Comunicação, tendo fornecido como fonte de dados 49 revistas científicas da área com 9864 documentos e 12 bibliotecas digitais de teses e dissertações com 1961 documentos. Os resultados apontam que os movimentos geradores e constituintes das redes sociais em nossos dois planos de análise são fortemente determinados por uma racionalidade característica da política científica do campo da Comunicação e da ciência de modo geral.The scientific production of an area of knowledge appears in different formats and is available in a distributed mainly through journals, proceedings, theses, dissertations and other typical formats used by the scientific community for the systematization of his speech. A federation of digital libraries offers an information architecture that aims to facilitate the aggregation of different types of documents available, facilitating access to those documents and their metadata descriptors, forming thus real structures to support the development of research and analysis of scientific documents that circulate through there. The analysis of social networks has proven an important subject of research in the area of Information Science and in recent decades have been appropriate even in a preliminary way by the Brazilian scientific community. As a way to increase knowledge and experimentation with the use of social network analysis and identify his potential analytical, the objective of this work was use network analysis to map the patterns , trends and connectivity strategies between two planes of relation between researchers: co-authoring of documents from scientific journals and participation in defenses of theses and dissertations. Furthermore, we seek to map the social and political causes of network patterns identified, highlighting a critical use of structural and dynamic indicators. We use as case Univerciencia.org federated digital library, a library specialized in the field of Communication Sciences and provided as a source of data collected 49 scientific journals in the area with 9864 documents and 12 digital libraries of theses and dissertations with 1961 documents. The results show that the generative movements and constituents of social networks in our two levels of analysis are strongly determined by a characteristic rationality of science policy in the field of communication and science in general.
13
Bhogi, Keerthana. "Two New Applications of Tensors to Machine Learning for Wireless Communications." Thesis, Virginia Tech, 2021. http://hdl.handle.net/10919/104970.
Full textAbstract:
With the increasing number of wireless devices and the phenomenal amount of data that is being generated by them, there is a growing interest in the wireless communications community to complement the traditional model-driven design approaches with data-driven machine learning (ML)-based solutions. However, managing the large-scale multi-dimensional data to maintain the efficiency and scalability of the ML algorithms has obviously been a challenge. Tensors provide a useful framework to represent multi-dimensional data in an integrated manner by preserving relationships in data across different dimensions. This thesis studies two new applications of tensors to ML for wireless communications where the tensor structure of the concerned data is exploited in novel ways.
The first contribution of this thesis is a tensor learning-based low-complexity precoder codebook design technique for a full-dimension multiple-input multiple-output (FD-MIMO) system with a uniform planar antenna (UPA) array at the transmitter (Tx) whose channel distribution is available through a dataset. Represented as a tensor, the FD-MIMO channel is further decomposed using a tensor decomposition technique to obtain an optimal precoder which is a function of Kronecker-Product (KP) of two low-dimensional precoders, each corresponding to the horizontal and vertical dimensions of the FD-MIMO channel. From the design perspective, we have made contributions in deriving a criterion for optimal product precoder codebooks using the obtained low-dimensional precoders. We show that this product codebook design problem is an unsupervised clustering problem on a Cartesian Product Grassmann Manifold (CPM), where the optimal cluster centroids form the desired codebook. We further simplify this clustering problem to a $K$-means algorithm on the low-dimensional factor Grassmann manifolds (GMs) of the CPM which correspond to the horizontal and vertical dimensions of the UPA, thus significantly reducing the complexity of precoder codebook construction when compared to the existing codebook learning techniques.
The second contribution of this thesis is a tensor-based bandwidth-efficient gradient communication technique for federated learning (FL) with convolutional neural networks (CNNs). Concisely, FL is a decentralized ML approach that allows to jointly train an ML model at the server using the data generated by the distributed users coordinated by a server, by sharing only the local gradients with the server and not the raw data. Here, we focus on efficient compression and reconstruction of convolutional gradients at the users and the server, respectively. To reduce the gradient communication overhead, we compress the sparse gradients at the users to obtain their low-dimensional estimates using compressive sensing (CS)-based technique and transmit to the server for joint training of the CNN. We exploit a natural tensor structure offered by the convolutional gradients to demonstrate the correlation of a gradient element with its neighbors. We propose a novel prior for the convolutional gradients that captures the described spatial consistency along with its sparse nature in an appropriate way. We further propose a novel Bayesian reconstruction algorithm based on the Generalized Approximate Message Passing (GAMP) framework that exploits this prior information about the gradients. Through the numerical simulations, we demonstrate that the developed gradient reconstruction method improves the convergence of the CNN model.Master of Science
The increase in the number of wireless and mobile devices have led to the generation of massive amounts of multi-modal data at the users in various real-world applications including wireless communications. This has led to an increasing interest in machine learning (ML)-based data-driven techniques for communication system design. The native setting of ML is {em centralized} where all the data is available on a single device. However, the distributed nature of the users and their data has also motivated the development of distributed ML techniques. Since the success of ML techniques is grounded in their data-based nature, there is a need to maintain the efficiency and scalability of the algorithms to manage the large-scale data. Tensors are multi-dimensional arrays that provide an integrated way of representing multi-modal data. Tensor algebra and tensor decompositions have enabled the extension of several classical ML techniques to tensors-based ML techniques in various application domains such as computer vision, data-mining, image processing, and wireless communications. Tensors-based ML techniques have shown to improve the performance of the ML models because of their ability to leverage the underlying structural information in the data. In this thesis, we present two new applications of tensors to ML for wireless applications and show how the tensor structure of the concerned data can be exploited and incorporated in different ways. The first contribution is a tensor learning-based precoder codebook design technique for full-dimension multiple-input multiple-output (FD-MIMO) systems where we develop a scheme for designing low-complexity product precoder codebooks by identifying and leveraging a tensor representation of the FD-MIMO channel. The second contribution is a tensor-based gradient communication scheme for a decentralized ML technique known as federated learning (FL) with convolutional neural networks (CNNs), where we design a novel bandwidth-efficient gradient compression-reconstruction algorithm that leverages a tensor structure of the convolutional gradients. The numerical simulations in both applications demonstrate that exploiting the underlying tensor structure in the data provides significant gains in their respective performance criteria.
14
Zhao, Qiwei. "Federated Learning with Heterogeneous Challenge." Thesis, The University of Sydney, 2022. https://hdl.handle.net/2123/27399.
Full textAbstract:
Federated learning allows the training of a model from the distributed data of many clients under the orchestration of a central server. With the increasing concern on privacy, federated learning draws great attention from both academia and industry. However, the heterogeneous challenges introduced by natural characters of federated learning settings significantly degrade the performance of federated learning methods. Specifically, these heterogeneous challenges include the heterogeneous data challenges and the heterogeneous scenario challenges.
Data heterogeneous challenges mean the significant differences between the datasets of numerous users. In federated learning, the data is stored separately on many distanced clients, causing these challenges. In addition, the heterogeneous scenario challenges refer to the differences between the devices participating in federated learning. Furthermore, the suitable models vary among the different scenarios. However, many existing federated learning methods use a single global model for all the devices' scenarios, which is not optimal for these two challenges.
We first propose a novel federated learning framework called local union in federated learning (LU-FL) to address these challenges. LU-FL incorporates the hierarchical knowledge distillation mechanism that effectively transfers knowledge among different models. So, LU-FL can enable any number of models to be used on each client. Allocating the specially designed models to different clients can mitigate the adverse effects caused by these challenges. At the same time, it can further improve the accuracy of the output models.
Extensive experimental results over several popular datasets demonstrate the effectiveness of our proposed method. It can effectively reduce the harmful effects of heterogeneous challenges, improving the accuracy of the final output models and the adaptability of the clients to various scenarios. So, it lets federated learning methods be applied in more diverse scenarios.
Keywords: federated learning, neural networks, knowledge distillation, computer vision
15
Lluch-Ariet, Magí. "Contributions to efficient and secure exchange of networked clinical data : the MOSAIC system." Doctoral thesis, Universitat Politècnica de Catalunya, 2016. http://hdl.handle.net/10803/388037.
Full textAbstract:
The understanding of certain data often requires the collection of similar data from different places to be analysed and interpreted. Multi-agent systems (MAS), interoperability standards (DICOM, HL7 or EN13606) and clinical Ontologies, are facilitating data interchange among different clinical centres around the world. However, as more and more data becomes available, and more heterogeneous this data gets, the task of accessing and exploiting the large number of distributed repositories to extract useful knowledge becomes increasingly complex. Beyond the existing networks and advances for data transfer, data sharing protocols to support multilateral agreements are useful to exploit the knowledge of distributed Data Warehouses. The access to a certain data set in a federated Data Warehouse may be constrained by the requirement to deliver another specific data set. When bilateral agreements between two nodes of a network are not enough to solve the constraints for accessing to a certain data set, multilateral agreements for data exchange can be a solution.
The research carried out in this PhD Thesis comprises the design and implementation of a Multi-Agent System for multilateral exchange agreements of clinical data, and evaluate how those multilateral agreements increase the percentage of data collected by a single node from the total amount of data available in the network. Different strategies to reduce the number of messages needed to achieve an agreement are also considered.
The results show that with this collaborative sharing scenario the percentage of data collected dramatically improve from bilateral agreements to multilateral ones, up to reach almost all data available in the network.
16
JIANG, JIN. "Social distributed content caching in federated residential networks." Doctoral thesis, Politecnico di Torino, 2013. http://hdl.handle.net/11583/2506271.
Full textAbstract:
This work addresses the need for content sharing and backup in household equipped with a home gateway that stores, tags and manages the data collected by the home users. Our solution leverages the interaction between remote gateways in a social way, i.e., by exploiting the users’ social networking information, so that caching recipients are those gateways whose users are most likely to be interested in accessing the shared content. We formulate this problem as a Budgeted Maximum Coverage (BMC) problem and we numerically compute the optimal content caching solution. We then propose a low-complexity, distributed heuristic algorithm and use simulation in a synthetic social network scenario to show that the final content placement among “friendly” gateways well approximates the optimal solution under different network settings.
17
Waugh, Daniel. "Oceania Football Confederation the impact of affiliate disaffiliation on the inter-organizational dynamics of a federated network : a dissertation submitted to Auckland University of Technology in partial fulfilment of the requirements for the degree of Master of Business (MBus), 2009 /." Click here to access this resource online, 2009. http://hdl.handle.net/10292/796.
Full text
18
Ariyattu, Resmi. "Towards federated social infrastructures for plug-based decentralized social networks." Thesis, Rennes 1, 2017. http://www.theses.fr/2017REN1S031/document.
Full textAbstract:
Dans cette thèse, nous abordons deux problèmes soulevés par les systèmes distribués décentralisés - le placement de réseaux logiques de façon compatible avec le réseau physique sous-jacent et la construction de cohortes d'éditeurs pour dans les systèmes d'édition collaborative. Bien que les réseaux logiques (overlay networks) été largement étudiés, la plupart des systèmes existant ne prennent pas ou prennent mal en compte la topologie du réseau physique sous-jacent, alors que la performance de ces systèmes dépend dans une grande mesure de la manière dont leur topologie logique exploite la localité présente dans le réseau physique sur lequel ils s'exécutent. Pour résoudre ce problème, nous proposons dans cette thèse Fluidify, un mécanisme décentralisé pour le déploiement d'un réseau logique sur une infrastructure physique qui cherche à maximiser la localité du déploiement. Fluidify utilise une stratégie double qui exploite à la fois les liaisons logiques d'un réseau applicatif et la topologie physique de son réseau sous-jacent pour aligner progressivement l'une avec l'autre. Le protocole résultant est générique, efficace, évolutif et peut améliorer considérablement les performances de l'ensemble. La deuxième question que nous abordons traite des plates-formes d'édition collaborative. Ces plates-formes permettent à plusieurs utilisateurs distants de contribuer simultanément au même document. Seuls un nombre limité d'utilisateurs simultanés peuvent être pris en charge par les éditeurs actuellement déployés. Un certain nombre de solutions pair-à-pair ont donc été proposées pour supprimer cette limitation et permettre à un grand nombre d'utilisateurs de collaborer sur un même document sans aucune coordination centrale. Ces plates-formes supposent cependant que tous les utilisateurs d'un système éditent le même jeu de document, ce qui est peu vraisemblable. Pour ouvrir la voie à des systèmes plus flexibles, nous présentons, Filament, un protocole décentralisé de construction de cohorte adapté aux besoins des grands éditeurs collaboratifs. Filament élimine la nécessité de toute table de hachage distribuée (DHT) intermédiaire et permet aux utilisateurs travaillant sur le même document de se retrouver d'une manière rapide, efficace et robuste en générant un champ de routage adaptatif autour d'eux-mêmes. L'architecture de Filament repose sur un ensemble de réseaux logiques auto-organisées qui exploitent les similarités entre jeux de documents édités par les utilisateurs. Le protocole résultant est efficace, évolutif et fournit des propriétés bénéfiques d'équilibrage de charge sur les pairs impliquésIn this thesis, we address two issues in the area of decentralized distributed systems: network-aware overlays and collaborative editing. Even though network overlays have been extensively studied, most solutions either ignores the underlying physical network topology, or uses mechanisms that are specific to a given platform or applications. This is problematic, as the performance of an overlay network strongly depends on the way its logical topology exploits the underlying physical network. To address this problem, we propose Fluidify, a decentralized mechanism for deploying an overlay network on top of a physical infrastructure while maximizing network locality. Fluidify uses a dual strategy that exploits both the logical links of an overlay and the physical topology of its underlying network to progressively align one with the other. The resulting protocol is generic, efficient, scalable and can substantially improve network overheads and latency in overlay based systems. The second issue that we address focuses on collaborative editing platforms. Distributed collaborative editors allow several remote users to contribute concurrently to the same document. Only a limited number of concurrent users can be supported by the currently deployed editors. A number of peer-to-peer solutions have therefore been proposed to remove this limitation and allow a large number of users to work collaboratively. These decentralized solution assume however that all users are editing the same set of documents, which is unlikely to be the case. To open the path towards more flexible decentralized collaborative editors, we present Filament, a decentralized cohort-construction protocol adapted to the needs of large-scale collaborative editors. Filament eliminates the need for any intermediate DHT, and allows nodes editing the same document to find each other in a rapid, efficient and robust manner by generating an adaptive routing field around themselves. Filament's architecture hinges around a set of collaborating self-organizing overlays that utilizes the semantic relations between peers. The resulting protocol is efficient, scalable and provides beneficial load-balancing properties over the involved peers
19
Backstad, Sebastian. "Federated Averaging Deep Q-NetworkA Distributed Deep Reinforcement Learning Algorithm." Thesis, Umeå universitet, Institutionen för datavetenskap, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-149637.
Full textAbstract:
In the telecom sector, there is a huge amount of rich data generated every day. This trend will increase with the launch of 5G networks. Telco companies are interested in analyzing their data to shape and improve their core businesses. However, there can be a number of limiting factors that prevents them from logging data to central data centers for analysis. Some examples include data privacy, data transfer, network latency etc. In this work, we present a distributed Deep Reinforcement Learning (DRL) method called Federated Averaging Deep Q-Network (FADQN), that employs a distributed hierarchical reinforcement learning architecture. It utilizes gradient averaging to decrease communication cost. Privacy concerns are also satisfied by training the agent locally and only sending aggregated information to the centralized server. We introduce two versions of FADQN: synchronous and asynchronous. Results on the cart-pole environment show 80 times reduction in communication without any significant loss in performance. Additionally, in case of asynchronous approach, we see a great improvement in convergence.
20
Basnayake, Mudiyanselage V. (Vishaka). "Federated learning for enhanced sensor reliability of automated wireless networks." Master's thesis, University of Oulu, 2019. http://jultika.oulu.fi/Record/nbnfioulu-201908142761.
Full textAbstract:
Abstract. Autonomous mobile robots working in-proximity humans and objects are becoming frequent and thus, avoiding collisions becomes important to increase the safety of the working environment. This thesis develops a mechanism to improve the reliability of sensor measurements in a mobile robot network taking into the account of inter-robot communication and costs of faulty sensor replacements. In this view, first, we develop a sensor fault prediction method utilizing sensor characteristics. Then, network-wide cost capturing sensor replacements and wireless communication is minimized subject to a sensor measurement reliability constraint. Tools from convex optimization are used to develop an algorithm that yields the optimal sensor selection and wireless information communication policy for aforementioned problem. Under the absence of prior knowledge on sensor characteristics, we utilize observations of sensor failures to estimate their characteristics in a distributed manner using federated learning. Finally, extensive simulations are carried out to highlight the performance of the proposed mechanism compared to several state-of-the-art methods.
21
Knight-Williams, Alex. "Time to Next Flow Classification in Mobile Networks with Federated Learning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-287183.
Full textAbstract:
Understanding traffic dynamics and user demand in a cellular network is essential for effective resource management, which in turn improves the network’s energy and cost efficiency. This thesis focuses on the task of classifying the time until the arrival of the next flow at a user level in a real network traffic data set. A range of machine learning and deep learning techniques are applied to this task, some of which are incorporated into a federated learning framework that ensures data privacy. The results demonstrate that the long short-term memory (LSTM) performs best on this task, although good performance can also be achieved with models of lower complexity. Furthermore, models developed through federated learning achieve comparable performance to those trained on centralised data.Att förstå trafikdynamik i ett cellulärt nätverk är fundamentalt för effektiv resurshantering, vilket förbättrar nätverkets energi- och kostnadseffektivitet. Denna uppsats fokuserar på att klassificera ankomsttid för nästa flöde på användarnivå i ett dataset innehållande nätverkstrafik från en operatör. En uppsättning maskininlärnings- och djupinlärningstekniker har applicerats, varav några har inkorporerats i ett ramverk för federerad inlärning som säkerställer användares dataintegritet. Resultaten visar att ett long short-term memory (LSTM) presterar bäst på denna uppgift, även om bra prestanda även kan uppnås med modeller av lägre komplexitet. Vidare så har modeller som tränats med hjälp av federerad inlärning jämförbar prestanda med de tränade på centraliserat data.
22
Mestoukirdi, Mohamad. "Reliable and Communication-Efficient Federated Learning for Future Intelligent Edge Networks." Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS432.
Full textAbstract:
Dans le domaine des futurs réseaux sans fil 6G, l'intégration de la périphérie intelligente grâce à l'avènement de l'IA représente un bond en avant considérable, promettant des avancées révolutionnaires en matière de communication sans fil. Cette intégration favorise une synergie harmonieuse, capitalisant sur le potentiel collectif de ces technologies transformatrices. Au cœur de cette intégration se trouve le rôle de l'apprentissage fédéré, un paradigme d'apprentissage décentralisé qui préserve la confidentialité des données tout en exploitant l'intelligence collective des appareils interconnectés. Dans la première partie de la thèse, nous nous attaquons au problème de l'hétérogénéité statistique dans l'apprentissage fédéré, qui découle des distributions de données divergentes entre les ensembles de données des dispositifs. Plutôt que d'entraîner un modèle unique conventionnel, qui donne souvent de mauvais résultats avec des données non identifiées, nous proposons un ensemble de règles centrées sur l'utilisateur qui produisent des modèles personnalisés adaptés aux objectifs de chaque utilisateur. Pour atténuer la surcharge de communication prohibitive associée à l'apprentissage d'un modèle personnalisé distinct pour chaque utilisateur, les utilisateurs sont répartis en groupes sur la base de la similarité de leurs objectifs. Cela permet l'apprentissage collectif de modèles personnalisés spécifiques à la cohorte. En conséquence, le nombre total de modèles personnalisés formés est réduit. Cette réduction diminue la consommation de ressources sans fil nécessaires à la transmission des mises à jour de modèles sur des canaux sans fil à bande passante limitée. Dans la deuxième partie, nous nous concentrons sur l'intégration des dispositifs à distance de l'IdO dans la périphérie intelligente en exploitant les véhicules aériens sans pilote en tant qu'orchestrateur d'apprentissage fédéré. Alors que des études antérieures ont largement exploré le potentiel des drones en tant que stations de base volantes ou relais dans les réseaux sans fil, leur utilisation pour faciliter l'apprentissage de modèles est encore un domaine de recherche relativement nouveau. Dans ce contexte, nous tirons parti de la mobilité des drones pour contourner les conditions de canal défavorables dans les zones rurales et établir des terrains d'apprentissage pour les dispositifs IoT distants. Cependant, les déploiements de drones posent des défis en termes de planification et de conception de trajectoires. À cette fin, une optimisation conjointe de la trajectoire du drone, de l'ordonnancement du dispositif et de la performance d'apprentissage est formulée et résolue à l'aide de techniques d'optimisation convexe et de la théorie des graphes. Dans la troisième partie de cette thèse, nous jetons un regard critique sur la surcharge de communication imposée par l'apprentissage fédéré sur les réseaux sans fil. Bien que les techniques de compression telles que la quantification et la sparsification des mises à jour de modèles soient largement utilisées, elles permettent souvent d'obtenir une efficacité de communication au prix d'une réduction de la performance du modèle. Pour surmonter cette limitation, nous utilisons des réseaux aléatoires sur-paramétrés pour approximer les réseaux cibles par l'élagage des paramètres plutôt que par l'optimisation directe. Il a été démontré que cette approche ne nécessite pas la transmission de plus d'un seul bit d'information par paramètre du modèle. Nous montrons que les méthodes SoTA ne parviennent pas à tirer parti de tous les avantages possibles en termes d'efficacité de la communication en utilisant cette approche. Nous proposons une fonction de perte régularisée qui prend en compte l'entropie des mises à jour transmises, ce qui se traduit par des améliorations notables de l'efficacité de la communication et de la mémoire lors de l'apprentissage fédéré sur des dispositifs périphériques, sans sacrifier la précisionIn the realm of future 6G wireless networks, integrating the intelligent edge through the advent of AI signifies a momentous leap forward, promising revolutionary advancements in wireless communication. This integration fosters a harmonious synergy, capitalizing on the collective potential of these transformative technologies. Central to this integration is the role of federated learning, a decentralized learning paradigm that upholds data privacy while harnessing the collective intelligence of interconnected devices. By embracing federated learning, 6G networks can unlock a myriad of benefits for both wireless networks and edge devices. On one hand, wireless networks gain the ability to exploit data-driven solutions, surpassing the limitations of traditional model-driven approaches. Particularly, leveraging real-time data insights will empower 6G networks to adapt, optimize performance, and enhance network efficiency dynamically. On the other hand, edge devices benefit from personalized experiences and tailored solutions, catered to their specific requirements. Specifically, edge devices will experience improved performance and reduced latency through localized decision-making, real-time processing, and reduced reliance on centralized infrastructure. In the first part of the thesis, we tackle the predicament of statistical heterogeneity in federated learning stemming from divergent data distributions among devices datasets. Rather than training a conventional one-model-fits-all, which often performs poorly with non-IID data, we propose user-centric set of rules that produce personalized models tailored to each user objectives. To mitigate the prohibitive communication overhead associated with training distinct personalized model for each user, users are partitioned into clusters based on their objectives similarity. This enables collective training of cohort-specific personalized models. As a result, the total number of personalized models trained is reduced. This reduction lessens the consumption of wireless resources required to transmit model updates across bandwidth-limited wireless channels. In the second part, our focus shifts towards integrating IoT remote devices into the intelligent edge by leveraging unmanned aerial vehicles as a federated learning orchestrator. While previous studies have extensively explored the potential of UAVs as flying base stations or relays in wireless networks, their utilization in facilitating model training is still a relatively new area of research. In this context, we leverage the UAV mobility to bypass the unfavorable channel conditions in rural areas and establish learning grounds to remote IoT devices. However, UAV deployments poses challenges in terms of scheduling and trajectory design. To this end, a joint optimization of UAV trajectory, device scheduling, and the learning performance is formulated and solved using convex optimization techniques and graph theory. In the third and final part of this thesis, we take a critical look at thecommunication overhead imposed by federated learning on wireless networks. While compression techniques such as quantization and sparsification of model updates are widely used, they often achieve communication efficiency at the cost of reduced model performance. We employ over-parameterized random networks to approximate target networks through parameter pruning rather than direct optimization to overcome this limitation. This approach has been demonstrated to require transmitting no more than a single bit of information per model parameter. We show that SoTA methods fail to capitalize on the full attainable advantages in terms of communication efficiency using this approach. Accordingly, we propose a regularized loss function which considers the entropy of transmitted updates, resulting in notable improvements to communication and memory efficiency during federated training on edge devices without sacrificing accuracy
23
Díaz, González Fernando. "Federated Learning for Time Series Forecasting Using LSTM Networks: Exploiting Similarities Through Clustering." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-254665.
Full textAbstract:
Federated learning poses a statistical challenge when training on highly heterogeneous sequence data. For example, time-series telecom data collected over long intervals regularly shows mixed fluctuations and patterns. These distinct distributions are an inconvenience when a node not only plans to contribute to the creation of the global model but also plans to apply it on its local dataset. In this scenario, adopting a one-fits-all approach might be inadequate, even when using state-of-the-art machine learning techniques for time series forecasting, such as Long Short-Term Memory (LSTM) networks, which have proven to be able to capture many idiosyncrasies and generalise to new patterns. In this work, we show that by clustering the clients using these patterns and selectively aggregating their updates in different global models can improve local performance with minimal overhead, as we demonstrate through experiments using realworld time series datasets and a basic LSTM model.Federated Learning utgör en statistisk utmaning vid träning med starkt heterogen sekvensdata. Till exempel så uppvisar tidsseriedata inom telekomdomänen blandade variationer och mönster över längre tidsintervall. Dessa distinkta fördelningar utgör en utmaning när en nod inte bara ska bidra till skapandet av en global modell utan även ämnar applicera denna modell på sin lokala datamängd. Att i detta scenario införa en global modell som ska passa alla kan visa sig vara otillräckligt, även om vi använder oss av de mest framgångsrika modellerna inom maskininlärning för tidsserieprognoser, Long Short-Term Memory (LSTM) nätverk, vilka visat sig kunna fånga komplexa mönster och generalisera väl till nya mönster. I detta arbete visar vi att genom att klustra klienterna med hjälp av dessa mönster och selektivt aggregera deras uppdateringar i olika globala modeller kan vi uppnå förbättringar av den lokal prestandan med minimala kostnader, vilket vi demonstrerar genom experiment med riktigt tidsseriedata och en grundläggande LSTM-modell.
24
Langelaar, Johannes, and Mattsson Adam Strömme. "Federated Neural Collaborative Filtering for privacy-preserving recommender systems." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-446913.
Full textAbstract:
In this thesis a number of models for recommender systems are explored, all using collaborative filtering to produce their recommendations. Extra focus is put on two models: Matrix Factorization, which is a linear model and Multi-Layer Perceptron, which is a non-linear model. With an additional purpose of training the models without collecting any sensitive data from the users, both models were implemented with a learning technique that does not require the server's knowledge of the users' data, called federated learning. The federated version of Matrix Factorization is already well-researched, and has proven not to protect the users' data at all; the data is derivable from the information that the users communicate to the server that is necessary for the learning of the model. However, on the federated Multi-Layer Perceptron model, no research could be found. In this thesis, such a model is therefore designed and presented. Arguments are put forth in support of the privacy preservability of the model, along with a proof of the user data not being analytically derivable for the central server. In addition, new ways to further put the protection of the users' data on the test are discussed. All models are evaluated on two different data sets. The first data set contains data on ratings of movies and is called MovieLens 1M. The second is a data set that consists of anonymized fund transactions, provided by the Swedish bank SEB for this thesis. Test results suggest that the federated versions of the models can achieve similar recommendation performance as their non-federated counterparts.
25
Li, Yuntao. "Federated Learning for Time Series Forecasting Using Hybrid Model." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-254677.
Full textAbstract:
Time Series data has become ubiquitous thanks to affordable edge devices and sensors. Much of this data is valuable for decision making. In order to use these data for the forecasting task, the conventional centralized approach has shown deficiencies regarding large data communication and data privacy issues. Furthermore, Neural Network models cannot make use of the extra information from the time series, thus they usually fail to provide time series specific results. Both issues expose a challenge to large-scale Time Series Forecasting with Neural Network models. All these limitations lead to our research question:Can we realize decentralized time series forecasting with a Federated Learning mechanism that is comparable to the conventional centralized setup in forecasting performance?In this work, we propose a Federated Series Forecasting framework, resolving the challenge by allowing users to keep the data locally, and learns a shared model by aggregating locally computed updates. Besides, we design a hybrid model to enable Neural Network models utilizing the extra information from the time series to achieve a time series specific learning. In particular, the proposed hybrid outperforms state-of-art baseline data-central models with NN5 and Ericsson KPI data. Meanwhile, the federated settings of purposed model yields comparable results to data-central settings on both NN5 and Ericsson KPI data. These results together answer the research question of this thesis.Tidseriedata har blivit allmänt förekommande tack vare överkomliga kantenheter och sensorer. Mycket av denna data är värdefull för beslutsfattande. För att kunna använda datan för prognosuppgifter har den konventionella centraliserade metoden visat brister avseende storskalig datakommunikation och integritetsfrågor. Vidare har neurala nätverksmodeller inte klarat av att utnyttja den extra informationen från tidsserierna, vilket leder till misslyckanden med att ge specifikt tidsserierelaterade resultat. Båda frågorna exponerar en utmaning för storskalig tidsserieprognostisering med neurala nätverksmodeller. Alla dessa begränsningar leder till vår forskningsfråga:Kan vi realisera decentraliserad tidsserieprognostisering med en federerad lärningsmekanism som presterar jämförbart med konventionella centrala lösningar i prognostisering?I det här arbetet föreslår vi ett ramverk för federerad tidsserieprognos som löser utmaningen genom att låta användaren behålla data lokalt och lära sig en delad modell genom att aggregera lokalt beräknade uppdateringar. Dessutom utformar vi en hybrid modell för att möjliggöra neurala nätverksmodeller som kan utnyttja den extra informationen från tidsserierna för att uppnå inlärning av specifika tidsserier. Den föreslagna hybrida modellen presterar bättre än state-of-art centraliserade grundläggande modeller med NN5och Ericsson KPIdata. Samtidigt ger den federerade ansatsen jämförbara resultat med de datacentrala ansatserna för både NN5och Ericsson KPI-data. Dessa resultat svarar tillsammans på forskningsfrågan av denna avhandling.
26
Leconte, Louis. "Compression and federated learning : an approach to frugal machine learning." Electronic Thesis or Diss., Sorbonne université, 2024. http://www.theses.fr/2024SORUS107.
Full textAbstract:
Les appareils et outils “intelligents” deviennent progressivement la norme, la mise en œuvre d'algorithmes basés sur des réseaux neuronaux artificiels se développant largement. Les réseaux neuronaux sont des modèles non linéaires d'apprentissage automatique avec de nombreux paramètres qui manipulent des objets de haute dimension et obtiennent des performances de pointe dans divers domaines, tels que la reconnaissance d'images, la reconnaissance vocale, le traitement du langage naturel et les systèmes de recommandation.Toutefois, l'entraînement d'un réseau neuronal sur un appareil à faible capacité de calcul est difficile en raison de problèmes de mémoire, de temps de calcul ou d'alimentation. Une approche naturelle pour simplifier cet entraînement consiste à utiliser des réseaux neuronaux quantifiés, dont les paramètres et les opérations utilisent des primitives efficaces à faible bit. Cependant, l'optimisation d'une fonction sur un ensemble discret en haute dimension est complexe et peut encore s'avérer prohibitive en termes de puissance de calcul. C'est pourquoi de nombreuses applications modernes utilisent un réseau d'appareils pour stocker des données individuelles et partager la charge de calcul. Une nouvelle approche a été proposée, l'apprentissage fédéré, qui prend en compte un environnement distribué : les données sont stockées sur des appareils différents et un serveur central orchestre le processus d'apprentissage sur les divers appareils.Dans cette thèse, nous étudions différents aspects de l'optimisation (stochastique) dans le but de réduire les coûts énergétiques pour des appareils potentiellement très hétérogènes. Les deux premières contributions de ce travail sont consacrées au cas des réseaux neuronaux quantifiés. Notre première idée est basée sur une stratégie de recuit : nous formulons le problème d'optimisation discret comme un problème d'optimisation sous contraintes (où la taille de la contrainte est réduite au fil des itérations). Nous nous sommes ensuite concentrés sur une heuristique pour la formation de réseaux neuronaux profonds binaires. Dans ce cadre particulier, les paramètres des réseaux neuronaux ne peuvent avoir que deux valeurs. Le reste de la thèse s'est concentré sur l'apprentissage fédéré efficace. Suite à nos contributions développées pour l'apprentissage de réseaux neuronaux quantifiés, nous les avons intégrées dans un environnement fédéré. Ensuite, nous avons proposé une nouvelle technique de compression sans biais qui peut être utilisée dans n'importe quel cadre d'optimisation distribuée basé sur le gradient. Nos dernières contributions abordent le cas particulier de l'apprentissage fédéré asynchrone, où les appareils ont des vitesses de calcul et/ou un accès à la bande passante différents. Nous avons d'abord proposé une contribution qui repondère les contributions des dispositifs distribués. Dans notre travail final, à travers une analyse détaillée de la dynamique des files d'attente, nous proposons une amélioration significative des bornes de complexité fournies dans la littérature sur l'apprentissage fédéré asynchrone.En résumé, cette thèse présente de nouvelles contributions au domaine des réseaux neuronaux quantifiés et de l'apprentissage fédéré en abordant des défis critiques et en fournissant des solutions innovantes pour un apprentissage efficace et durable dans un environnement distribué et hétérogène. Bien que les avantages potentiels soient prometteurs, notamment en termes d'économies d'énergie, il convient d'être prudent car un effet rebond pourrait se produire“Intelligent” devices and tools are gradually becoming the standard, as the implementation of algorithms based on artificial neural networks is experiencing widespread development. Neural networks consist of non-linear machine learning models that manipulate high-dimensional objects and obtain state-of-the-art performances in various areas, such as image recognition, speech recognition, natural language processing, and recommendation systems.However, training a neural network on a device with lower computing capacity can be challenging, as it can imply cutting back on memory, computing time or power. A natural approach to simplify this training is to use quantized neural networks, whose parameters and operations use efficient low-bit primitives. However, optimizing a function over a discrete set in high dimension is complex, and can still be prohibitively expensive in terms of computational power. For this reason, many modern applications use a network of devices to store individual data and share the computational load. A new approach, federated learning, considers a distributed environment: Data is stored on devices and a centralized server orchestrates the training process across multiple devices.In this thesis, we investigate different aspects of (stochastic) optimization with the goal of reducing energy costs for potentially very heterogeneous devices. The first two contributions of this work are dedicated to the case of quantized neural networks. Our first idea is based on an annealing strategy: we formulate the discrete optimization problem as a constrained optimization problem (where the size of the constraint is reduced over iterations). We then focus on a heuristic for training binary deep neural networks. In this particular framework, the parameters of the neural networks can only have two values. The rest of the thesis is about efficient federated learning. Following our contributions developed for training quantized neural network, we integrate them into a federated environment. Then, we propose a novel unbiased compression technique that can be used in any gradient based distributed optimization framework. Our final contributions address the particular case of asynchronous federated learning, where devices have different computational speeds and/or access to bandwidth. We first propose a contribution that reweights the contributions of distributed devices. Then, in our final work, through a detailed queuing dynamics analysis, we propose a significant improvement to the complexity bounds provided in the literature onasynchronous federated learning.In summary, this thesis presents novel contributions to the field of quantized neural networks and federated learning by addressing critical challenges and providing innovative solutions for efficient and sustainable learning in a distributed and heterogeneous environment. Although the potential benefits are promising, especially in terms of energy savings, caution is needed as a rebound effect could occur
27
Hark, Rhaban Simon [Verfasser], Ralf [Akademischer Betreuer] Steinmetz, and Andreas [Akademischer Betreuer] Mauthe. "Monitoring Federated Softwarized Networks: Approaches for Efficient and Collaborative Data Collection in Large-Scale Software-Defined Networks / Rhaban Simon Hark ; Ralf Steinmetz, Andreas Mauthe." Darmstadt : Universitäts- und Landesbibliothek Darmstadt, 2019. http://d-nb.info/1196792550/34.
Full text
28
Ricco, Christophe. "Soluzione Android-based per il Discovery e l'Accesso a Servizi fra Reti Locali Federate." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2019.
Find full textAbstract:
Negli ultimi anni abbiamo assistito alla rapida proliferazione di dispositivi sempre più piccoli in termini di dimensioni ma con delle capacità computazionali sempre più elevate. Con il numero di questi dispositivi in continua crescita, unita alle tecnologie di comunicazione sempre più economiche è nata quella che viene chiamata Internet of Things. La caratteristica principale per far parte di Internet of Things è quella di essere in grado di comunicare con gli altri. Ad oggi le maggiore tecnologie utilizzate sono Wifi e LTE ma necessitano di una infrastruttura esistente sul territorio. Al fine di poter arginare il vincolo delle infrastrutture, sono stati implementati una serie di middleware che permettono di costruire delle particolari reti di oggetti connessi, che in letteratura vengono definite \textbf{MANET}, esempio Google Nearby. Poichè il contesto in qui vengono applicati è altamente dinamico e volatile, è fondamentale avere la possibilità di scoprire la presenza di servizi all'interno della rete, senza bisogno di nessun tipo di conoscenza. Il problema però è che non tutti i middleware offrono questa funzionalità e molte volte i normali servizi di discovery non possono essere utilizzati. Lo studio fatto in questo elaborato è mirato a cercare una soluzione a basso costo che permette di abilitare un sistema di discovery su reti federate fornendo ricerca e accesso ai servizi. La soluzione proposta è stata quella di sfruttare un servizio di discovery come DNS-SD (chiamato anche Bonjour) per scoprire i servizi in una rete locale basata su Wifi e comunicarli ad altre reti attraverso una comunicazione Wifi Direct. Come per i vari middleware, l'ipotesi alla base del funzionamento del servizio è la vicinanza geografica delle reti, in modo da poter creare un ponte tra di esse attraverso una connessione P2P.
29
Subramanya, Tejas. "Autonomic Management and Orchestration Strategies in MEC-Enabled 5G Networks." Doctoral thesis, Università degli studi di Trento, 2021. http://hdl.handle.net/11572/320883.
Full textAbstract:
5G and beyond mobile network technology promises to deliver unprecedented ultra-low latency and high data rates, paving the way for many novel applications and services. Network Function Virtualization (NFV) and Multi-access Edge Computing (MEC) are two technologies expected to play a vital role in achieving ambitious Quality of Service requirements of such applications. While NFV provides flexibility by enabling network functions to be dynamically deployed and inter-connected to realize Service Function Chains (SFC), MEC brings the computing capability to the mobile network's edges, thus reducing latency and alleviating the transport network load. However, adequate mechanisms are needed to meet the dynamically changing network service demands (i.e., in single and multiple domains) and optimally utilize the network resources while ensuring that the end-to-end latency requirement of services is always satisfied. In this dissertation work, we break the problem into three separate stages and present the solutions for each one of them.Firstly, we apply Artificial Intelligence (AI) techniques to drive NFV resource orchestration in MEC-enabled 5G architectures for single and multi-domain scenarios. We propose three deep learning approaches to perform horizontal and vertical Virtual Network Function (VNF) auto-scaling: (i) Multilayer Perceptron (MLP) classification and regression (single-domain), (ii) Centralized Artificial Neural Network (ANN), centralized Long-Short Term Memory (LSTM) and centralized Convolutional Neural Network-LSTM (CNN-LSTM) (single-domain), and (iii) Federated ANN, federated LSTM and federated CNN-LSTM (multi-domain). We evaluate the performance of each of these deep learning models trained over a commercial network operator dataset and investigate the pros and cons of different approaches for VNF auto-scaling. For the first approach, our results show that both MLP classifier and MLP regressor models have strong predicting capability for auto-scaling. However, MLP regressor outperforms MLP classifier in terms of accuracy. For the second approach (one-step prediction), CNN-LSTM performs the best for the QoS-prioritized objective and LSTM performs the best for the cost-prioritized objective. For the second approach (multi-step prediction), the encoder-decoder CNN-LSTM model outperforms the encoder-decoder LSTM model for both QoS and Cost prioritized objectives. For the third approach, both federated LSTM and federated CNN-LSTM models perform equally better than the federated ANN model. It was also noted that in general federated learning approaches performs poorly compared to centralized learning approaches. Secondly, we employ Integer Linear Programming (ILP) techniques to formulate and solve a joint user association and SFC placement problem, where each SFC represents a service requested by a user with end-to-end latency and data rate requirements. We also develop a comprehensive end-to-end latency model considering radio delay, backhaul network delay and SFC processing delay for 5G mobile networks. We evaluated the proposed model using simulations based on real-operator network topology and real-world latency values. Our results show that the average end-to-end latency reduces significantly when SFCs are placed at the ME hosts according to their latency and data rate demands. Furthermore, we propose an heuristic algorithm to address the issue of scalability in ILP, that can solve the above association/mapping problem in seconds rather than hours.Finally, we introduce lightMEC - a lightweight MEC platform for deploying mobile edge computing functionalities which allows hosting of low-latency and bandwidth-intensive applications at the network edge. Measurements conducted over a real-life test demonstrated that lightMEC could actually support practical MEC applications without requiring any change to existing mobile network nodes' functionality in the access and core network segments. The significant benefits of adopting the proposed architecture are analyzed based on a proof-of-concept demonstration of the content caching use case. Furthermore, we introduce the AI-driven Kubernetes orchestration prototype that we implemented by leveraging the lightMEC platform and assess the performance of the proposed deep learning models (from stage 1) in an experimental setup. The prototype evaluations confirm the simulation results achieved in stage 1 of the thesis.
30
Griffier, Romain. "Intégration et utilisation secondaire des données de santé hospitalières hétérogènes : des usages locaux à l'analyse fédérée." Electronic Thesis or Diss., Bordeaux, 2024. http://www.theses.fr/2024BORD0479.
Full textAbstract:
Les données issues du soin peuvent être utilisées pour des finalités autres que celles pour lesquelles elles ont été collectées initialement : c’est l’utilisation secondaire des données de santé. Dans le contexte hospitalier, afin de lever les verrous de l’utilisation secondaire des données de santé (verrous liés aux données et verrous organisationnels), une stratégie classique consiste à mettre en place un Entrepôt de Données de Santé (EDS). Dans le cadre de cette thèse, trois contributions à l’EDS du CHU de Bordeaux sont décrites. Premièrement, une méthode d’alignement des data éléments de biologie numérique basée sur les instances et conforme aux règles de protection des données à caractère personnel est présentée, avec une F-mesure à 0,850, permettant de réduire l’hétérogénéité sémantique des données. Ensuite, une adaptation du modèle d’intégration des données cliniques d’i2b2 est proposée pour assurer la persistance des données d’un EDS dans une base de données NoSQL, Elasticsearch. Cette implémentation a été évaluée sur la base de données de l’EDS du CHU de Bordeaux et retrouve des performances améliorées en termes de stockage et de temps de requêtage, par rapport à une base de données relationnelle. Enfin, une présentation de l’environnement EDS du CHU de Bordeaux est réalisée, avec la description d’un premier EDS dédié aux usages locaux et qui peut être exploité en autonomie par les utilisateurs finaux (i2b2), et d’un second EDS, dédié aux réseaux fédérés (OMOP) permettant notamment la participation au réseau fédéré DARWIN-EUHealthcare data can be used for purposes other than those for which it was initially collected: this is the secondary use of health data. In the hospital context, to overcome the obstacles to secondary use of healthcaree data (data and organizational barriers), a classic strategy is to set up Clinical Data Warehouses (CDWs). This thesis describes three contributions to the Bordeaux University Hospital’s CDW. Firstly, an instance-based, privacy-preserving, method for mapping numerical biology data elements is presented, with an F-measure of 0,850, making it possible to reduce the semantic heterogeneity of data. Next, an adaptation of the i2b2 clinical data integration model is proposed to enable CDW data persistence in a NoSQL database, Elasticsearch. This implementation has been evaluated on the Bordeaux University Hospital’s CDW, showing improved performance in terms of storage and query time, compared with a relational database. Finally, the Bordeaux University Hospital’s CDW environment is presented, with the description of a first CDW dedicated to local uses that can be used autonomously by end users (i2b2), and a second CDW dedicated to federated networks (OMOP) enabling participation in the DARWIN-EU federated network
31
Chafaa, Irched. "Machine learning for beam alignment in mmWave networks." Electronic Thesis or Diss., université Paris-Saclay, 2021. http://www.theses.fr/2021UPASG044.
Full textAbstract:
Pour faire face à la croissance exponentielle du trafic des données mobiles, une solution possible est d'exploiter les larges bandes spectrales disponibles dans la partie millimétrique du spectre électromagnétique. Cependant, le signal transmis est fortement atténué, impliquant une portée de propagation limitée et un faible nombre des trajets de propagation (canal parcimonieux). Par conséquent, des faisceaux directifs doivent être utilisés pour focaliser l'énergie du signal transmis vers son utilisateur et compenser les pertes de propagation. Ces faisceaux ont besoin d'être dirigés convenablement pour garantir la fiabilité du lien de communication. Ceci représente le problème d'alignement des faisceaux pour les systèmes de communication à onde millimétrique. En effet, les faisceaux de l'émetteur et du récepteur doivent être constamment ajustés et alignés pour combattre les conditions de propagation difficiles de la bande millimétrique. De plus, les techniques d'alignement des faisceaux doivent prendre en compte la mobilité des utilisateurs et la dynamique imprévisible du réseau. Ceci mène à un fort coût de signalisation et d'entraînement qui impacte les performances des réseaux. Dans la première partie de cette thèse, nous reformulons le problème d'alignement des faisceaux en utilisant les bandits manchots (ou multi-armed bandits), pertinents dans le cas d'une dynamique du réseau imprévisibles et arbitraire (non-stationnaire ou même antagoniste). Nous proposons des méthodes en ligne et adaptatives pour aligner indépendamment les faisceaux des deux nœuds du lien de communication en utilisant seulement un seul bit de feedback. En se basant sur l'algorithme des poids exponentiels (EXP3) et le caractère parcimonieux du canal à onde millimétrique, nous proposons une version modifiée de l'algorithme original (MEXP3) avec des garanties théoriques en fonction du regret asymptotique. En outre, pour un horizon du temps fini, notre borne supérieure du regret est plus serrée que celle de l'algorithme EXP3, indiquant une meilleure performance en pratique. Nous introduisons également une deuxième modification qui utilise les corrélations temporelles entre des choix successifs des faisceaux dans une nouvelle technique d'alignement des faisceaux (NBT-MEXP3). Dans la deuxième partie de cette thèse, des outils de l'apprentissage profond sont examinés pour choisir des faisceaux dans un lien point d'accès -- utilisateur. Nous exploitons l'apprentissage profond non supervisé pour utiliser l'information des canaux au-dessous de 6 GHz afin de prédire des faisceaux dans la bande millimétrique; cette fonction canal-faisceau complexe est apprise en utilisant des données non-annotés du dataset DeepMIMO. Nous discutons aussi le choix d'une taille optimale pour le réseau de neurones en fonction du nombre des antennes de transmission et de réception au point d'accès. De plus, nous étudions l'impact de la disponibilité des données d'entraînement et introduisons une approche basée sur l'apprentissage fédéré pour prédire des faisceaux dans un réseau à plusieurs liens en partageant uniquement les paramètres des réseaux de neurones entrainés localement (et non pas les données locales). Nous envisageons les méthodes synchrones et asynchrones de l'approche par apprentissage fédéré. Nos résultats numériques montrent le potentiel de notre approche particulièrement au cas où les données d'entrainement sont peu abondantes ou imparfaites (bruitées). Enfin, nous comparons nos méthodes basées sur l'apprentissage profond avec celles de la première partie. Les simulations montrent que le choix d'une méthode convenable pour aligner les faisceaux dépend de la nature de l'application et présente un compromis entre le débit obtenu et la complexité du calculTo cope with the ever increasing mobile data traffic, an envisioned solution for future wireless networks is to exploit the large available spectrum in the millimeter wave (mmWave) band. However, communicating at these high frequencies is very challenging as the transmitted signal suffers from strong attenuation, which leads to a limited propagation range and few multipath components (sparse mmWave channels). Hence, highly-directional beams have to be employed to focus the signal energy towards the intended user and compensate all those losses. Such beams need to be steered appropriately to guarantee a reliable communication link. This represents the so called beam alignment problem where the beams of the transmitter and the receiver need to be constantly aligned. Moreover, beam alignment policies need to support devices mobility and the unpredicted dynamics of the network, which result in significant signaling and training overhead affecting the overall performance. In the first part of the thesis, we formulate the beam alignment problem via the adversarial multi-armed bandit framework, which copes with arbitrary network dynamics including non-stationary or adversarial components. We propose online and adaptive beam alignment policies relying only on one-bit feedback to steer the beams of both nodes of the communication link in a distributed manner. Building on the well-known exponential weights algorithm (EXP3) and by exploiting the sparse nature of mmWave channels, we propose a modified policy (MEXP3), which comes with optimal theoretical guarantees in terms of asymptotic regret. Moreover, for finite horizons, our regret upper-bound is tighter than that of the original EXP3 suggesting better performance in practice. We then introduce an additional modification that accounts for the temporal correlation between successive beams and propose another beam alignment policy (NBT-MEXP3). In the second part of the thesis, deep learning tools are investigated to select mmWave beams in an access point -- user link. We leverage unsupervised deep learning to exploit the channel knowledge at sub-6 GHz and predict beamforming vectors in the mmWave band; this complex channel-beam mapping is learned via data issued from the DeepMIMO dataset and lacking the ground truth. We also show how to choose an optimal size of our neural network depending on the number of transmit and receive antennas at the access point. Furthermore, we investigate the impact of training data availability and introduce a federated learning (FL) approach to predict the beams of multiple links by sharing only the parameters of the locally trained neural networks (and not the local data). We investigate both synchronous and asynchronous FL methods. Our numerical simulations show the high potential of our approach, especially when the local available data is scarce or imperfect (noisy). At last, we compare our proposed deep learning methods with reinforcement learning methods derived in the first part. Simulations show that choosing an appropriate beam steering method depends on the target application and is a tradeoff between rate performance and computational complexity
32
Hao, Jialin. "Machine learning for road active safety in vehicular networks." Electronic Thesis or Diss., Institut polytechnique de Paris, 2024. http://www.theses.fr/2024IPPAS003.
Full textAbstract:
Cette thèse porte sur le développement d'une manœuvre d'aide au changement de voie (lane Change Assistance, LCA) sûre et efficace dans le contexte des réseaux de véhicules assistés par drones (Drone Assisted Vehicular Network, DAVN). En effet, les changements de voie contribuent de manière significative aux accidents de la route, nécessitant des solutions efficaces au sein des réseaux routiers. Les LCA stratégies actuelles établies sur l'apprentissage par renforcement profond (Deep Reinforcement Learning, DRL) sont limitées par les informations locales sur les véhicules, négligeant une vue globale, comme des conditions de circulation. Pour résoudre ce problème, les véhicules aériens sans pilote (Unmanned Aerial Vehicles, UAVs), ou drones, présentent une extension prometteuse des services de réseau automobile grâce à leur mobilité, capacités informatiques et liaisons de communication en visibilité directe (Line-if-Sight, LoS) avec les véhicules routiers. Dans un premier temps, nous faisons une étude bibliographique sur LCA au sein du DAVN, mettant en évidence le potentiel des drones pour améliorer la sécurité routière. Les approches LCA existantes s'appuient principalement sur des informations locales sur les véhicules et ne prennent pas en compte l'état global du trafic. Afin de réduire cette limitation, nous proposons le GL-DEAR : joint global and local drone-assisted lane change platform based on Deep-Q Network (DQN) with a dynamic reward function, for LCA with drones' assistance. La plateforme proposée se compose de trois modules : route à risques aléatoires et véhicules d'urgence ; acquisition et traitement des données ; prise de décision de changement de voie en temps réel. La manœuvre de changement de voie est basée sur un Deep Q-Network avec des fonctions de récompense dynamiques. Plus précisément, nous adoptons les modèles de changement de voie authentiques basés sur l'ensemble de données NGSIM pour les véhicules routiers ordinaires afin de recréer les comportements de changement de voie du monde réel dans les simulations. Les résultats numériques démontrent la capacité de la plateforme à réaliser des trajets sans collision sur des autoroutes à risque avec des véhicules d'urgence. Dans un deuxième temps, nous identifions un manque de calibrage de la fréquence de mise à jour globale des algorithmes d'apprentissage fédéré (Federated Learning, FL) et l'absence d'évaluation approfondie du délai de traitement au niveau du drone. Nous proposons donc un cadre d'apprentissage par renforcement fédéré (FRL) assisté par drone, DAFL. Ce cadre permet un apprentissage coopératif entre les véhicules de l'ego en appliquant FL. Il comprend un algorithme d'agrégation de modèles global basé sur la réputation du client et une analyse complète du délai de bout en bout (End-to-End, E2E) au niveau du drone. Plus précisément, la fréquence globale de mise à jour est ajustée dynamiquement en fonction des mesures de sécurité routière et de la consommation énergétique des drones, ce qui donne des résultats efficaces dans les simulations. Dans la troisième étape, nous concevons l'algorithme DOP-T pour optimiser les trajectoires des drones dans les réseaux de véhicules dynamiques. Cet algorithme vise à équilibrer la consommation énergétique des drones et la sécurité routière. Nous fournissons un état de l'art complet des techniques existantes de planification de trajectoire de drones. Ensuite, sur la base de la modélisation du délai E2E du véhicule et de la modélisation de la consommation d'énergie du drone. Dans la seconde étape, nous formons un modèle d'apprentissage par renforcement hors ligne (Offline-Reinforcement Learning, ORL) pour éviter une formation en ligne consommatrice d'énergie. Les résultats de la simulation démontrent une réduction significative de la consommation d'énergie des drones et du délai E2E du véhicule à l'aide du modèle entraînéThis thesis focuses on the development of a safe and efficient LCA maneuver in the context of drone-assisted vehicle networks (DAVN). In fact, lane change maneuvers contribute significantly to road accidents, requiring effective solutions within road networks. Current lane change assistance (LCA) strategies relying solely on deep reinforcement learning (DRL) are limited by local vehicle information, neglecting a global view of traffic conditions. To address this problem, unmanned aerial vehicles (UAVs), or drones, present a promising extension of automotive network services due to their mobility, computing capabilities, and line-of-sight (LoS) communications links with road vehicles. In the first step, we conduct a literature review on LCA within DAVN, highlighting the potential of drones to enhance road safety. Existing LCA approaches predominantly rely on local vehicle information and fail to consider overall traffic states. To address this limitation, we propose the GL-DEAR: joint global and local drone-assisted lane change platform based on Deep-Q Network (DQN) with a dynamic reward function, for LCA with drones' assistance. The proposed platform consists of three modules: road with random risks and emergency vehicles; data file acquisition and processing; and real-time lane change decision-making. The lane change maneuver is based on a Deep Q-Network with dynamic reward functions. Specifically, we adopt the authentic NGSIM dataset-based lane change models for ordinary road vehicles to recreate real world lane change behaviors in the simulations. Numerical results demonstrate the platform's ability to achieve collision-free trips on risky highways with emergency vehicles. In the second step, we identify a lack of calibration for the global update frequency in FL algorithms and the absence of thorough drone-level processing delay assessment. To this end, we propose the drone assisted Federated Reinforcement Learning (FRL)-based LCA framework, DAFL. This framework enables cooperative learning between ego vehicles by applying Federated Learning (FL). It includes a client reputation-based global model aggregation algorithm and a comprehensive analysis of End-to-End (E2E) delay at the drone. Specifically, the global update frequency is dynamically adjusted according to road safety measurements and drone energy consumption, yielding efficient results in simulations. In the third step, we devise the DOP-T algorithm for optimizing drone trajectories in dynamic vehicular networks. This algorithm aims to balance drone energy consumption and road safety. We provide a comprehensive state-of-the-art review of the existing drone trajectory planning techniques. Then, based on the vehicle E2E delay modeling and the drone energy consumption modeling in the second step, we train a Offline Reinforcement Learning (ORL) model to avoid power-consuming online training. Simulation results demonstrate a significant reduction in drone energy consumption and vehicle E2E delay using the trained model
33
Roth, Ana Lúcia. "Proposição de esquema conceitual para a governança de redes de cooperação federadas." Universidade do Vale do Rio dos Sinos, 2011. http://www.repositorio.jesuita.org.br/handle/UNISINOS/5187.
Full textAbstract:
Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2016-04-18T18:05:24Z
No. of bitstreams: 1
ANA LÚCIA ROTH_.pdf: 1643482 bytes, checksum: 4e4cda2780cc7255c2502065566263c6 (MD5)Made available in DSpace on 2016-04-18T18:05:24Z (GMT). No. of bitstreams: 1 ANA LÚCIA ROTH_.pdf: 1643482 bytes, checksum: 4e4cda2780cc7255c2502065566263c6 (MD5) Previous issue date: 2011-08-30
Nenhuma
O presente trabalho versa sobre a governança de Redes de Cooperação Federadas, uma tipologia específica das Relações Interorganizacionais. O estudo buscou, motivado pelas lacunas teóricas observadas na literatura, identificar os elementos que compõem o sistema de governança das Redes de Cooperação Federadas e como eles podem ser projetados para reduzir os conflitos e os custos de agência nas Federações. Percebeu-se que os estudos que descrevem e analisam os elementos internos da Governança das RIs não chegam a abordar a separação da propriedade e do controle, característico das Redes de Cooperação Federadas. Autores que versam sobre a separação da propriedade e do controle não apresentam os elementos internos da Governança das RIs. Da mesma forma, os estudos não chegam a tratar como os elementos internos da governança (estrutura e mecanismos de governança) podem ser combinados para minimizar os conflitos e custos de agência. Sob a luz da Teoria da Agência, da Teoria da Representação e da Teoria Organizacional usada no contexto das RIs, chegou-se à proposição de três elementos do sistema de governança das Redes de Cooperação Federadas: (a) a estrutura de governança; (b) os mecanismos de governança instituídos para incentivar e controlar o comportamento dos agentes; (c) os mecanismos de governança instituídos para incentivar e controlar o comportamento dos principais. Para a validação empírica do modelo proposto, optou-se pela realização de um estudo qualitativo, de natureza exploratória, por meio do método de estudo de caso múltiplo. O estudo foi operacionalizado por meio de dois estudos de caso: um na Rede Âncora Brasil e o outro na Rede Construir. Os resultados permitem afirmar que, embora presentes, os elementos estruturais e mecanismos de governança sobre os principais e sobre os agentes apresentam características distintas. Há, em um extremo, uma Federação com um maior nível de instâncias de decisão, maior número de atividades desenvolvidas pela Federação, o que tende a levar a uma maior centralização, formalização, padronização, coordenação e controle. No outro extremo, encontra-se uma Federação com poucas instâncias de decisão, menor número de atividades desenvolvidas pela Federação, o que tende a levar a uma descentralização, menor formalização, padronização, coordenação e controle. O estudo empírico levou à adição do elemento representação e participação na estrutura de governança das Redes de Cooperação Federadas. Os mecanismos internos e externos de governança, amplamente utilizados nas corporações também foram identificados no contexto das Redes de Cooperação Federadas. Constatou-se, entretanto, que o Conselho de Administração ou órgão correspondente é o mecanismo mais efetivo para redução dos custos e conflitos de agência. Entre as contribuições teóricas do trabalho, destaca-se a adoção de uma nova perspectiva do conceito de governança das Redes de Cooperação Federadas sob a luz da Teoria da Agência.
This study concerns Federated Cooperative Networks governance, a specific Interorganizational Relationship (IOR) typology. Motivated by a theoretical gap in this subject, we sought to identify the elements forming the Federated Cooperative Networks governance system and how they can be designed to reduce conflicts and agency costs in Federations. It was noticed that studies describing and analyzing internal IORs governance elements do not approach separation of ownership from control, characteristic of Federated Cooperative Networks. Authors who discuss separation of ownership from control don’t show the internal elements of IOR governance. And the studies don’t discuss how the internal governance elements (governance structure and mechanisms) might be combined to minimize conflicts and agency costs either. Based on the Agency Theory, Representation Theory and Organizational Theory employed in IORs context, we have proposed three elements for the Federated Cooperative Networks governance system: (a) a governance structure; (b) governance mechanisms instituted to encourage and control the behavior of agents; (c) governance mechanisms instituted to encourage and control the behavior of principals. For the empirical validation of these elements, we have chosen to perform a qualitative study, of exploratory nature, through the multiple case study method. The study was operated through two case studies: one at Rede Âncora Brasil and the other at Rede Construir. The results allow us to assert that, although present, structural elements and governance mechanisms over principals and agents show distinct features. On one extreme, there is a Federation with a higher level of decision instances, a higher number of activities developed by the Federation, which tend to lead to more centralization, formalization, standardization, coordination and control. On the other extreme, a Federation with little decision instances, and a lower number of activities developed by the Federation, which tend to lead to decentralization, less formalization, standardization, coordination and control. The empirical study added the element representation and participation in the Federated Cooperative Networks governance structure. The internal and external governance mechanisms, widely employed in corporations, were also identified in the Federated Cooperative Networks context. However, we have found that the Administrative Council, or corresponding agency, is the most effective mechanism in reducing agency costs and conflicts. Among the theoretical contributions of the study, adopting a new perspective in the concept of Federated Cooperative Networks governance under the light of the Agency theory can be stressed.
34
Toofanee, Mohammud Shaad Ally. "An innovative ecosystem based on deep learning : Contributions for the prevention and prediction of diabetes complications." Electronic Thesis or Diss., Limoges, 2023. https://aurore.unilim.fr/theses/nxfile/default/656b0a1f-2ff2-49c5-bb3e-f34704d6f6b0/blobholder:0/2023LIMO0107.pdf.
Full textAbstract:
En 2021, le diabète touchait environ 537 millions de personnes dans le monde. Ce chiffre devrait grimper à 643 millions d'ici 2030 et 783 millions d'ici 2045. Le diabète est une maladie métabolique persistante qui nécessite des soins et une gestion quotidiens continus. Le fardeau des maladies chroniques pèse lourdement sur les systèmes de santé lorsqu'il touche une partie substantielle de la population. De telles circonstances ont un impact négatif non seulement sur le bien-être général d'une grande partie de la population, mais contribuent également de manière significative aux dépenses de santé. Dans le contexte de Maurice, selon le rapport le plus récent de la Fédération Internationale du Diabète, la prévalence du diabète, en particulier du diabète de type 2 (T2D), était de 22,6 % de la population en 2021, avec des projections indiquant une hausse à 26,6 % d'ici 2045. Face à cette tendance alarmante, une évolution concomitante a été observée dans le domaine de la technologie, l'intelligence artificielle démontrant des capacités prometteuses dans les domaines de la médecine et de la santé. Cette thèse de doctorat entreprend l'exploration de l'intersection entre l'intelligence artificielle, plus précisément l’apprentissage profond, l'éducation, la prévention, et la gestion du diabète. Nous nous sommes d'abord concentrés sur l'exploration du potentiel de l'Intelligence Artificielle (IA) pour répondre à une complication fréquente du diabète : l'Ulcère du Pied Diabétique (DFU). Les DFU présentent un risque grave d'amputations des membres inférieurs, entraînant des conséquences graves. En réponse, nous avons proposé une solution innovante nommée DFU-HELPER. Cet outil permet de valider les protocoles de traitement administrés par les professionnels de la santé aux patients individuels atteints de DFU. L'évaluation initiale de l'outil a montré des résultats prometteurs, bien qu'un affinement further et des tests rigoureux soient impératifs. Les efforts collaboratifs avec les experts en santé publique seront essentiels pour évaluer l'efficacité pratique de l'outil dans des scénarios réels. Cette approche vise à combler le fossé entre les technologies IA et les interventions cliniques, avec pour objectif ultime d'améliorer la prise en charge des patients atteints de DFU. Notre recherche a également abordé les aspects critiques de la vie privée et de la confidentialité inhérents à la manipulation des données liées à la santé. Reconnaissant l'importance capitale de la protection des informations sensibles, nous avons appliqué une approche avancée d'apprentissage fédéré Peer-to-Peer à notre proposition pour l'outil DFU-Helper. Cette approche permet de traiter des données sensibles sans les transférer vers un serveur central, contribuant ainsi à créer un environnement de confiance et sécurisé pour la gestion des données de santé. Enfin, notre recherche s'est étendue au développement d'un agent conversationnel intelligent conçu pour fournir des informations et un soutien 24 heures sur 24 aux personnes atteintes de diabète. Dans la poursuite de cet objectif, la création d'un jeu de données approprié était essentielle. Dans ce contexte, nous avons utilisé des techniques de traitement du langage naturel pour sélectionner des données de qualité provenant de sources médias en ligne traitant du diabèteIn the year 2021, estimations indicated that approximately 537 million individuals were affected by diabetes, a number anticipated to escalate to 643 million by the year 2030 and further to 783 million by 2045. Diabetes, characterized as a persistent metabolic ailment, necessitates unceasing daily care and management. In the context of Mauritius, as per the most recent report by the International Diabetes Federation, the prevalence of diabetes, specifically Type 2 Diabetes (T2D), stood at 22.6% of the population in 2021, with projections indicating a surge to 26.6% by the year 2045. Amidst this alarming trend, a concurrent advancement has been observed in the realm of technology, with artificial intelligence techniques showcasing promising capabilities in the spheres of medicine and healthcare. This doctoral dissertation embarks on the exploration of the intersection between artificial intelligence and diabetes education, prevention, and management.We initially focused on exploring the potential of artificial intelligence (AI), more specifically, deep learning, to address a critical complication linked to diabetes – Diabetic Foot Ulcer (DFU). The emergence of DFU poses the grave risk of lower limb amputations, consequently leading to severe socio-economic repercussions. In response, we put forth an innovative solution named DFU-HELPER. This tool serves as a preliminary measure for validating the treatment protocols administered by healthcare professionals to individual patients afflicted by DFU. The initial assessment of the proposed tool has exhibited promising performance characteristics, although further refinement and rigorous testing are imperative. Collaborative efforts with public health experts will be pivotal in evaluating the practical efficacy of the tool in real-world scenarios. This approach seeks to bridge the gap between AI technologies and clinical interventions, with the ultimate goal of improving the management of patients with DFU.Our research also addressed the critical aspects of privacy and confidentiality inherent in handling health-related data. Acknowledging the extreme importance of safeguarding sensitive information, we delved into the realm of Peer-to-Peer Federated Learning. This investigation specifically found application in our proposal for the DFU-Helper tool discussed earlier. By exploring this advanced approach, we aimed to ensure that the implementation of our technology aligns with privacy standards, thereby fostering a trustworthy and secure environment for healthcare data management.Finally, our research extended to the development of an intelligent conversational agent designed to offer round-the-clock support for individuals seeking information about diabetes. In pursuit of this goal, the creation of an appropriate dataset was paramount. In this context, we leveraged Natural Language Processing techniques to curate data from online media sources focusing on diabetes-related content
35
Yu-Ting, Lai, and 賴瑀庭. "Design and Development of a Resource Control Framework for Federated Network Testbed." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/56556564524408044895.
Full textAbstract:
碩士國立成功大學
電腦與通信工程研究所
103
With the rapid development of cloud computing and virtualization techniques, the needs for applications to use multiple resources distributed in different network testbeds, a.k.a resource sites have also grown greatly. With this demands, an improved and scalable Resource Control Framework is required to replace the traditional management orchestration used by each resource site. This adaptive Resource Control Framework is supposed to offer a mechanism to dynamically allocate and join physical resources located in varied sites according requirements from applications. This thesis designs a global Resource Control Framework to provide a platform as Networked Testbed over the Software Defined Network for users to deploy their applications in an isolated network environment. The architecture of RCF realizes the concept of network resource pool to assemble multiple resource sites to share their resources with each other. A distributed inter-domain Virtual Network Embedding method is devised to deploy an application across different resource sites. Moreover, a new LAN community based division method is used to improve the performance of inter-domain and intra-domain VNE algorithm. The simulation results show the better embedding revenue and higher application acceptance ratio. In addition, a VLAN tag translation mechanism is adopted to stitch the application virtual network between resource sites. This thesis implements RCF in two resource sites at NCKU EE building and CHIMEI building respectively, forming the first federated, Network Testbed overlay the production network in Taiwan.
36
Lin, Bor-Hsuin, and 林伯勳. "A study on Spatial Data Transformation Format and Network Communication Under Federated GIS." Thesis, 1996. http://ndltd.ncl.edu.tw/handle/00809396105064601402.
Full textAbstract:
碩士國立成功大學
測量工程學系
84
Among the governmental agencies, there exist many land-related information systems well operated at the present moment. These systems reveal common characteristics. That is they are developed solely according to the requirements of the agencies developing them. Each system is designed and tuned to fully meet the particular needs of the specific agency. When it comes to users from other agencies that have to access the systems, many problems arise. For instance, when the cross-agencies data are needed, users have to manually collect data of different formats from systems in different agencies. This is inconvenient as well as inefficient. If we could manage to connect systems in related agencies, and have the compatibility problems solved, we would be able to access data across systems on-line directly through a single system. There are many feasible approaches to cope with the integration of multiple systems. In this thesis, we'll refer to one of the concepts of multiple databases (or multi-databases), the federated architecture, and discuss the feasibility of its application to the land-related information systems. The issues of integrating multiple systems are of wide range. However, we'll give a general description to the characteristics of the federated system, and put our efforts on selecting a data transformation format to meet the system's integration requirements. Finally, we'll follow the concept of the federated architecture to implement a simplified system that has the network communication established and provides on-line access to data from different systems. We shall focus on the implementation of the prototype of the architecture mentioned above with short discussions on how data can be integrated.
37
Esteves, Leonardo Galveias. "Federated Learning for IoT Edge Computing: An Experimental Study." Master's thesis, 2022. http://hdl.handle.net/10316/99424.
Full textAbstract:
Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e TecnologiaOs dados gerados por anualmente rondam os 40 trilhões de gigabytes. Este aumento significativo de dados todos os anos trás a necessidade de assegurar a proteção de informação sensível. A Inteligência Artificial tem vindo a melhorar cada vez mais os seus resultados, apresentando modelos capazes de responder rigorosamente em áreas de atuação críticas, por exemplo, medicina, veículos autónomos, robótica, etc. Estes algoritmos precisam de enormes quantidades de dados disponíveis para otimizarem ao máximo a sua resposta perante todos a sua área de operação.Surgiu a necessidade de continuar a melhorar estes algoritmos mantendo a privacidade e confidencialidade dos dados utilizados.Desta forma, foi criado o conceito de Federated Learning. O Federated Learning permite continuar a treinar algoritmos de Machine Learning sem partilhar os dados utilizados para a convergência do modelo. O Federated Learning apresenta apresenta algumas similaridades com o Distributed Learning. Em ambos os conceitos o treino é distribuido, no entanto o Federated Learning descentraliza também os dados de forma a manter a informação privada.O objetivo desta dissertação passa por explorar o conceito de Federated Learning, assim como comparar diretamente este conceito com o Machine Learning centralizado. Para tal, é mostrada a arquitetura necessária para a construção de uma solução federada. Este documento apresenta ainda resultados obtidos com soluções federadas tanto em ambiente de simulação como numa implementação em ambiente real. Finalmente, é também apresentado um ponto de vista dos resultados obtidos e opções de otimização de uma solução com Federated Learning são discutidas.
The data generated annually is around 40 trillion gigabytes. This significant increase in data every year brings with it the need to ensure the protection of sensitive information. Artificial Intelligence has been improving its results more and more, presenting models capable of responding rigorously in critical areas, for example medicine, autonomous vehicles, robotics, etc. These algorithms need huge amounts of available data to optimize their response to all their area of operation.The urge to continue to improve these algorithms while maintaining the privacy and confidentiality of the data used emerged.Thus, the concept of Federated Learning was created. Federated Learning allows to continue training Machine Learning algorithms without sharing the data used for model convergence. Federated Learning has some similarities with Distributed Learning. In both concepts the training is distributed, however, Federated Learning also decentralizes the data in order to keep the information private.The objective of this dissertation is to explore the concept of Federated Learning, as well as to directly compare this concept with centralized Machine Learning. To this end, the architecture required to build a federated solution is analyzed in depth. This dissertation also presents results obtained with federated solutions in both simulation and real-world deployment. Finally, a viewpoint of the obtained results is also presented, and options for optimizing a solution with Federated Learning are discussed.
38
Thomas, Paul. "Server characterisation and selection for personal metasearch." Phd thesis, 2008. http://hdl.handle.net/1885/150244.
Full text
39
Eirinha, Tiago Filipe Rodrigues. "Extended Federated Social Networks in Environmental Sustainability." Master's thesis, 2017. http://hdl.handle.net/11110/1583.
Full textAbstract:
A enorme e rápida adesão ao uso das redes sociais, leva-nos a inferir que representam um dos mais relevantes
fenómenos ligados à internet e às Tecnologias de Informação e Comunicação (TIC) em geral. Na generalidade
dos casos, sempre que um comportamento assenta num rápido “crescimento”, um conjunto de questões serão
levantadas. As redes sociais federadas surgiram no seguimento deste fenómeno, procurando colmatar algumas
das lacunas identificadas nas redes sociais atuais, nomeadamente ao nível da portabilidade dos dados, da
identidade, da referenciação e da privacidade. Após uma análise cuidada, pensa-se que algo mais se pode
acoplar a este conceito.
Cada vez mais é essencial perceber o contexto em que as “coisas” acontecem e orientar o desenvolvimento de
aplicações no sentido de apresentarem alguma “consciência” ou sensibilidade ao contexto onde são utilizadas,
i.e., em vez de trabalharem de uma forma isolada, poderem de alguma forma colaborar entre si. Assim, é
possível usufruir de múltiplas aplicações, de diferentes tipos e fornecedores, a interoperarem entre elas e a
adaptarem-se ao contexto global, num dashboard comum. Se a isto acrescentarmos a possibilidade de as
aplicações serem partilhadas entre membros de uma rede, então teremos uma rede constituída por nodos com
capacidade de, autonomamente, reconfigurarem-se e adaptarem-se ao contexto necessário.
Existindo a sensibilidade ao contexto e, consequentemente, a deteção de necessidades, torna-se indispensável
a existência de mecanismos capazes de auxiliarem a obtenção de respostas (brokering) com opções viáveis,
permitindo resolver “problemas” de uma forma ágil, rápida e eficaz.
A eficácia na colaboração promove a co-decisão e a capacidade de comunicação humano-humano representa
a essência neste processo. Nas redes sociais atuais, a existência de múltiplos canais (meios ou serviços) de
comunicação representa uma mais-valia, mas a não integração desses mesmos canais, evidencia um sério
constrangimento, forçando os utilizadores a estarem presentes em várias plataformas (redes) de maneira a
atingir todas pessoas que desejam. É evidente nos dias de hoje a tendência destas plataformas em uniformizar
entre si formas de comunicação integradas (chats, videoconferências, outras). Existe assim e ainda, a
necessidade de unificar (ou melhor, integrar) a comunicação humano-humano, tornando-a mais próxima dos
atuais sistemas telefónicos em relação à sua arquitetura de funcionamento. Neste sistema existem múltiplos
provedores de serviço, mas o utilizador não necessita de se preocupar com o “canal” usado. Não está na sua
consciência se o recetor para quem pretende comunicar possui ou não um canal (serviço) igual.
Este trabalho propõe uma arquitetura de sistemas e desenvolve um protótipo onde se incorporam componentes
que tentam responder às extensões identificadas, nomeadamente uma arquitetura de interoperabilidade de
widgets, uma arquitetura comunicacional e um sistema de brokering.
40
(10725357), Siddharth Divi. "UNIFYING DISTILLATION WITH PERSONALIZATION IN FEDERATED LEARNING." Thesis, 2021.
Find full textAbstract:
Federated learning (FL) is a decentralized privacy-preserving learning technique in which clients learn a joint collaborative model through a central aggregator without sharing their data. In this setting, all clients learn a single common predictor (FedAvg), which does not generalize well on each client's local data due to the statistical data heterogeneity among clients. In this paper, we address this problem with PersFL, a discrete two-stage personalized learning algorithm. In the first stage, PersFL finds the optimal teacher model of each client during the FL training phase. In the second stage, PersFL distills the useful knowledge from optimal teachers into each user's local model. The teacher model provides each client with some rich, high-level representation that a client can easily adapt to its local model, which overcomes the statistical heterogeneity present at different clients. We evaluate PersFL on CIFAR-10 and MNIST datasets using three data-splitting strategies to control the diversity between clients' data distributions.
We empirically show that PersFL outperforms FedAvg and three state-of-the-art personalization methods, pFedMe, Per-FedAvg and FedPer on majority data-splits with minimal communication cost. Further, we study the performance of PersFL on different distillation objectives, how this performance is affected by the equitable notion of fairness among clients, and the number of required communication rounds. We also build an evaluation framework with the following modules: Data Generator, Federated Model Generation, and Evaluation Metrics. We introduce new metrics for the domain of personalized FL, and split these metrics into two perspectives: Performance, and Fairness. We analyze the performance of all the personalized algorithms by applying these metrics to answer the following questions: Which personalization algorithm performs the best in terms of accuracy across all the users?, and Which personalization algorithm is the fairest amongst all of them? Finally, we make the code for this work available at https://tinyurl.com/1hp9ywfa for public use and validation.
41
Hark, Rhaban Simon. "Monitoring Federated Softwarized Networks: Approaches for Efficient and Collaborative Data Collection in Large-Scale Software-Defined Networks." Phd thesis, 2019. https://tuprints.ulb.tu-darmstadt.de/9073/1/2019-08-15_Hark_Rhaban_Simon.pdf.
Full textAbstract:
The term Softwarized Networking encapsulates technologies that allow the use of software to program a communication network. These
technologies, predominantly Software-Defined Networking (SDN) and Network Functions Virtualization (NFV), have dominated the scientific interests of the networking community in the last decade. Leading companies already adopted SDN in large-scale deployments (e.g., Google’s B4 Project, Microsoft Azure). According to Cisco, 76% of all data centers will apply SDN by 2021. Along with a hand full of valuable advantages, the foundation of the success of Softwarized Networking lies in its flexibility. In the case of SDN, a logically centralized controller, denoted control-plane, uses software to dynamically change how the networking devices, denoted data-plane, handle traffic. This centralization tremendously eases the management process. With respect to network state monitoring, which is a cornerstone of network management and the basis for its adaptivity, SDN provides, in addition to the advantage of the by-design
centrally available knowledge, a set of new techniques to collect statistics from the networking devices.
The centralization of the control-plane quickly turned out to be only of logical nature and requires a physically distributed implementation to achieve scalability and reliability. Therefore, numerous distributed controller architectures have been proposed. Yet, the distribution of the control and, in line with this, the distribution of large-scale networks (e.g., one data center consists of a multitude of distributed sub-data centers) have not been considered in existing monitoring approaches. We believe there is a potential to increase the efficiency of monitoring when network parts collaborate.
In this thesis, we exploit this potential by developing monitoring approaches that utilize coordination and information exchange among collaborating SDN controllers. We create mechanisms to discover redundancy in the monitoring of shared resources and aggregate overlapping measurement tasks of different controllers whenever possible. Doing so, we substantially cut down the costs for monitoring, which is necessary for future networks that face a vast increase in load and dynamics. On top of this, we zoom into the statistic collection process in Softwarized Networks between controllers and the data-plane devices. Within that, we identify three not yet fully explored aspects, namely how, where, and which statistics to measure from the network. We propose novel methods for these aspects to collect information efficiently while limiting resource consumption. Extensive evaluations show that filtering irrelevant data can reduce the required measurement transmissions to a fraction and an intelligent measurement point placement requires only a small number of measurements compared to measuring the entire network, without affecting the accuracy.
42
LIN, CHIA-YEN, and 林佳延. "Application Study of Federated Unscented Kalman Filter on Dual-Band Infrared Search and Track Sensor Networks." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/926469.
Full textAbstract:
碩士育達科技大學
資訊管理所
106
In recent years, unmanned aerial photographing of flying robots has become increasingly popular, and accidents have occurred due to improper operation of personnel, and even shooting in military forbidden areas has been heard. In view of this, the government revised the relevant laws to regulate the prohibition of over flight. When enforcing the law, the current policy must be reported by the public or patrolled by relevant law enforcement agencies personnel to the relevant locations. In order to save time, manpower, cost, and control of electronic evidence, multiple stations including dual-band infrared search and track sensors are needed, and target tracking is performed through sensor network federated filter calculations. The method is developed to obtain noisy data of target in passive ranging and angles and then sends the measurement data to dedicated local processor (federated sub-filter). However in the sensor level, the target range and the angles of azimuth and elevation are measured by each dual-band infrared search and track sensor in the sphere coordinate system. Each local processor uses the Unscented Kalman Filter (UKF) to perform state estimation and finally transmits the information to the global processor for information fusion. The target state is estimated by UKF in the reference rectangular coordinate system. Then the UKF handles the recursion and update of the state mean vector and the error covariance matrix via the Unscented Transformations (UT), which makes the estimation algorithm closer to the nonlinear nature of the system. Finally, the state information of each local processor is transmitted to the global processor (federated master filter) to integrate these state estimations to a final state estimation for system output and information feedback. To test the effectiveness of the algorithm, Monte Carlo computer simulation technique is adopted. After repeated simulation experiments, the algorithm has excellent performance in tracking accuracy. It can be known from the simulation results that the algorithm proposed in this thesis is suitable for the tracking application of flying objects. Keywords: Federated Filter, Unscented Kalman Filter, Dual-Band Infrared Search and Track Sensor Networks.
43
Snopová, Karolína. "Činnost Evropské cyklistické federace jako součást evropské turistiky." Master's thesis, 2016. http://www.nusl.cz/ntk/nusl-342064.
Full textAbstract:
Title: Activities of the European Cyclists' Federation as a part of European tourism Objectives: The thesis is focused on cycling tourism in relation to European tourism, European Union policy and global policy of the world organizations. The aim is to present the activities of the European Cyclists' Federation, characterize the organization and analyze its biggest project - the development of the European network of long-distance cycling routes EuroVelo. This case study also shows the way of cooperation of a non-governmental organization, the European Union and world organizations in order to support the development of cycling tourism and fulfill the objectives of the European Union policy and global policy of world organizations. Methods: The information has been obtained by document analysis and by communication with the former Vice President of the European Cyclists' Federation and currently a member of the Executive Board of EuroVelo within the ECF. Results: Used methods have brought enough information to create complete description of the European Cyclists' Federation and its activities including the EuroVelo project. The result is a complex text that shows the importance of cycling (and thus activities of ECF) for the society and demonstrates the cooperation of organizations at all levels in...
44
Vu, Thanh Tung. "Collaborative processing and radio resource management for cloud-based radio access networks." Thesis, 2021. http://hdl.handle.net/1959.13/1430295.
Full textAbstract:
Research Doctorate - Doctor of Philosophy (PhD)Next-generation wireless communication systems are expected to deliver extreme improvements in throughput and energy efficiency, in an era of pervasive machine learning applications. Meeting the ambitious technical objectives set out in beyond-5G visions requires a collaborative effort among a large number of wireless access points. Leveraging advances in optimisation theory, this thesis studies collaborative radio resource management for the two most promising cloud-based wireless network architectures: Cloud Radio Access Networks (C-RANs) and Cell-Free Massive Multiple-Input-Multiple-Output (CFmMIMO) networks. In these networks, the significant advantages achieved through centralised baseband processing, spatial multiplexing and large antenna arrays make it possible to support distributed machine learning algorithms over the air. However, on top of the traditional issue of limited available radio resources, C-RANs and CFmMIMO architectures must resolve new issues arising from the limited-capacity fronthaul links that connect the access points with the central processing unit in the cloud. Furthermore, executing distributed machine learning algorithms over dynamic and unpredictable wireless links remains a major obstacle. The first contribution of this thesis is proposing novel resource allocation solutions to improve throughput and energy efficiency for C-RANs via edge baseband processing. Here, we propose that baseband signals be partially processed at the access points instead of being fully processed at the central processing unit. We show that in the best-case scenarios, our optimisation-based edge processing approach provides up to 50% energy efficiency gain over the existing centralised compression-based solutions. We further extend our results to full-duplex transmissions, and show that full-duplexing can improve the network throughput by almost 1.5 times while tripling the energy-efficiency gain, compared with half-duplex transmissions. In the second contribution, we consider a C-RAN content delivery network where each access point stores the frequently requested content in its local cache. Local caching has been shown to substantially reduce data traffic on the fronthaul links. We propose new transmission and resource allocation algorithms that are applicable to any caching scheme. Here, both multicasting and unicasting cases are analysed in detail. Simulation results show that for a given caching scheme, our proposals can deliver up to 20% network throughput gain and 14% energy efficiency gain. In the third contribution, we propose a novel communication scheme for CFmMIMO networks to support any federated learning (FL) framework. This scheme exploits the channel hardening property of massive MIMO, where the effective channel gains remain relatively unchanged during the large-scale channel coherence time. We propose that within this coherence time, one (instead of all) iteration of an FL process is executed, thus guaranteeing the stable operation of the whole FL process. In this communication scheme, we develop successive convex approximation algorithms that allocate radio resources to minimise the FL training time. Simulation results show that our proposals reduce the FL training time by more than half compared to the heuristic approaches. In the fourth contribution of this thesis, we extend the results in the third contribution to mitigate the `straggler effect' in wireless federated learning. Here, a user with unfavorable wireless link conditions may significantly slow down the entire FL process. We study new user selection methods for CFmMIMO networks and develop algorithms that minimise the FL training time. Significantly, a reduction of up to 63% in training time has been observed in our numerical results with practical parameter settings.