Dissertations / Theses on the topic 'Approaches to learning'
To see the other types of publications on this topic, follow the link: Approaches to learning.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Approaches to learning.'
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
Potari, Despina. "Learning approaches in mathematics." Thesis, University of Edinburgh, 1987. http://hdl.handle.net/1842/12130.
Full text
2
Hussein, Ahmed. "Deep learning based approaches for imitation learning." Thesis, Robert Gordon University, 2018. http://hdl.handle.net/10059/3117.
Full textAbstract:
Imitation learning refers to an agent's ability to mimic a desired behaviour by learning from observations. The field is rapidly gaining attention due to recent advances in computational and communication capabilities as well as rising demand for intelligent applications. The goal of imitation learning is to describe the desired behaviour by providing demonstrations rather than instructions. This enables agents to learn complex behaviours with general learning methods that require minimal task specific information. However, imitation learning faces many challenges. The objective of this thesis is to advance the state of the art in imitation learning by adopting deep learning methods to address two major challenges of learning from demonstrations. Firstly, representing the demonstrations in a manner that is adequate for learning. We propose novel Convolutional Neural Networks (CNN) based methods to automatically extract feature representations from raw visual demonstrations and learn to replicate the demonstrated behaviour. This alleviates the need for task specific feature extraction and provides a general learning process that is adequate for multiple problems. The second challenge is generalizing a policy over unseen situations in the training demonstrations. This is a common problem because demonstrations typically show the best way to perform a task and don't offer any information about recovering from suboptimal actions. Several methods are investigated to improve the agent's generalization ability based on its initial performance. Our contributions in this area are three fold. Firstly, we propose an active data aggregation method that queries the demonstrator in situations of low confidence. Secondly, we investigate combining learning from demonstrations and reinforcement learning. A deep reward shaping method is proposed that learns a potential reward function from demonstrations. Finally, memory architectures in deep neural networks are investigated to provide context to the agent when taking actions. Using recurrent neural networks addresses the dependency between the state-action sequences taken by the agent. The experiments are conducted in simulated environments on 2D and 3D navigation tasks that are learned from raw visual data, as well as a 2D soccer simulator. The proposed methods are compared to state of the art deep reinforcement learning methods. The results show that deep learning architectures can learn suitable representations from raw visual data and effectively map them to atomic actions. The proposed methods for addressing generalization show improvements over using supervised learning and reinforcement learning alone. The results are thoroughly analysed to identify the benefits of each approach and situations in which it is most suitable.
3
Effraimidis, Dimitros. "Computation approaches for continuous reinforcement learning problems." Thesis, University of Westminster, 2016. https://westminsterresearch.westminster.ac.uk/item/q0y82/computation-approaches-for-continuous-reinforcement-learning-problems.
Full textAbstract:
Optimisation theory is at the heart of any control process, where we seek to control the behaviour of a system through a set of actions. Linear control problems have been extensively studied, and optimal control laws have been identified. But the world around us is highly non-linear and unpredictable. For these dynamic systems, which don’t possess the nice mathematical properties of the linear counterpart, the classic control theory breaks and other methods have to be employed. But nature thrives by optimising non-linear and over-complicated systems. Evolutionary Computing (EC) methods exploit nature’s way by imitating the evolution process and avoid to solve the control problem analytically. Reinforcement Learning (RL) from the other side regards the optimal control problem as a sequential one. In every discrete time step an action is applied. The transition of the system to a new state is accompanied by a sole numerical value, the “reward” that designate the quality of the control action. Even though the amount of feedback information is limited into a sole real number, the introduction of the Temporal Difference method made possible to have accurate predictions of the value-functions. This paved the way to optimise complex structures, like the Neural Networks, which are used to approximate the value functions. In this thesis we investigate the solution of continuous Reinforcement Learning control problems by EC methodologies. The accumulated reward of such problems throughout an episode suffices as information to formulate the required measure, fitness, in order to optimise a population of candidate solutions. Especially, we explore the limits of applicability of a specific branch of EC, that of Genetic Programming (GP). The evolving population in the GP case is comprised from individuals, which are immediately translated to mathematical functions, which can serve as a control law. The major contribution of this thesis is the proposed unification of these disparate Artificial Intelligence paradigms. The provided information from the systems are exploited by a step by step basis from the RL part of the proposed scheme and by an episodic basis from GP. This makes possible to augment the function set of the GP scheme with adaptable Neural Networks. In the quest to achieve stable behaviour of the RL part of the system a modification of the Actor-Critic algorithm has been implemented. Finally we successfully apply the GP method in multi-action control problems extending the spectrum of the problems that this method has been proved to solve. Also we investigated the capability of GP in relation to problems from the food industry. These type of problems exhibit also non-linearity and there is no definite model describing its behaviour.
4
Chang, Yu-Han Ph D. Massachusetts Institute of Technology. "Approaches to multi-agent learning." Thesis, Massachusetts Institute of Technology, 2005. http://hdl.handle.net/1721.1/33932.
Full textAbstract:
Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2005.Includes bibliographical references (leaves 165-171).
Systems involving multiple autonomous entities are becoming more and more prominent. Sensor networks, teams of robotic vehicles, and software agents are just a few examples. In order to design these systems, we need methods that allow our agents to autonomously learn and adapt to the changing environments they find themselves in. This thesis explores ideas from game theory, online prediction, and reinforcement learning, tying them together to work on problems in multi-agent learning. We begin with the most basic framework for studying multi-agent learning: repeated matrix games. We quickly realize that there is no such thing as an opponent-independent, globally optimal learning algorithm. Some form of opponent assumptions must be necessary when designing multi-agent learning algorithms. We first show that we can exploit opponents that satisfy certain assumptions, and in a later chapter, we show how we can avoid being exploited ourselves. From this beginning, we branch out to study more complex sequential decision making problems in multi-agent systems, or stochastic games. We study environments in which there are large numbers of agents, and where environmental state may only be partially observable.
(cont.) In fully cooperative situations, where all the agents receive a single global reward signal for training, we devise a filtering method that allows each individual agent to learn using a personal training signal recovered from this global reward. For non-cooperative situations, we introduce the concept of hedged learning, a combination of regret-minimizing algorithms with learning techniques, which allows a more flexible and robust approach for behaving in competitive situations. We show various performance bounds that can be guaranteed with our hedged learning algorithm, thus preventing our agent from being exploited by its adversary. Finally, we apply some of these methods to problems involving routing and node movement in a mobilized ad-hoc networking domain.
by Yu-Han Chang.
Ph.D.
5
Flaherty, Drew. "Artistic approaches to machine learning." Thesis, Queensland University of Technology, 2020. https://eprints.qut.edu.au/200191/1/Drew_Flaherty_Thesis.pdf.
Full textAbstract:
This research is about how Artificial Intelligence and Machine Learning may impact creative practice. The thesis looks at various implementations and models related to the subject from different cultural and technical viewpoints. The project also provides experimental creative outcomes from my personal practice along with a qualitative study into attitudes and perspectives from other creative practitioners.
6
Yu, Kai. "Statistical Learning Approaches to Information Filtering." Diss., lmu, 2004. http://nbn-resolving.de/urn:nbn:de:bvb:19-25120.
Full text
7
Kashima, Hisashi. "Machine learning approaches for structured data." 京都大学 (Kyoto University), 2007. http://hdl.handle.net/2433/135953.
Full text
8
Chen, Zhe Haykin Simon S. "Stochastic approaches for correlation-based learning." *McMaster only, 2004.
Find full text
9
Boots, Byron. "Spectral Approaches to Learning Predictive Representations." Research Showcase @ CMU, 2012. http://repository.cmu.edu/dissertations/131.
Full textAbstract:
A central problem in artificial intelligence is to choose actions to maximize reward in a partially observable, uncertain environment. To do so, we must obtain an accurate environment model, and then plan to maximize reward. However, for complex domains, specifying a model by hand can be a time consuming process. This motivates an alternative approach: learning a model directly from observations. Unfortunately, learning algorithms often recover a model that is too inaccurate to support planning or too large and complex for planning to succeed; or, they require excessive prior domain knowledge or fail to provide guarantees such as statistical consistency. To address this gap, we propose spectral subspace identification algorithms which provably learn compact, accurate, predictive models of partially observable dynamical systems directly from sequences of action-observation pairs. Our research agenda includes several variations of this general approach: spectral methods for classical models like Kalman filters and hidden Markov models, batch algorithms and online algorithms, and kernel-based algorithms for learning models in high- and infinite-dimensional feature spaces. All of these approaches share a common framework: the model’s belief space is represented as predictions of observable quantities and spectral algorithms are applied to learn the model parameters. Unlike the popular EM algorithm, spectral learning algorithms are statistically consistent, computationally efficient, and easy to implement using established matrixalgebra techniques. We evaluate our learning algorithms on a series of prediction and planning tasks involving simulated data and real robotic systems.
10
Pellegrini, Giovanni. "Relational Learning approaches for Recommender Systems." Doctoral thesis, Università degli studi di Trento, 2021. http://hdl.handle.net/11572/318892.
Full textAbstract:
Learning on relational data is a relevant task in the machine learning community. Extracting information from structured data is a non-trivial task due to the combinatorial complexity of the domain and the necessity to construct methods that work on collections of values of different sizes rather than fixed representations. Relational data can naturally be interpreted as graphs, a class of flexible and expressive structures that can model data from diverse domains,from biology to social interactions. Graphs have been used in a huge variety of contexts, such as molecular modelling, social networks, image processing and recommendation systems. In this manuscript, we tackle some challenges in learning on relational data by developing new learning methodologies. Specifically, in our first contribution, we introduce a new class of metrics for relational data based on relational features extraction technique called Type ExtensionTrees. This class of metrics defines the (dis)similarity of two nodes in a graph by exploiting the nested structure of their relational neighborhood at different depth steps. In our second contribution, we developed a new strategy to collect the information of multisets of data values by introducing a new framework of learnable aggregators called Learning Aggregation Functions.We provide a detailed description of the methodologies and an extensive experimental evaluation on synthetic and real world data to assess the expressiveness of the proposed models. A particular focus is given to the application of these methods to the recommendation systems domain, exploring the combination of the proposed methods with recent techniques developed for Constructive Preference Elicitation and Group Recommendation tasks.
11
Fu, I.-Ping P. "Student Approaches to Learning Chinese Vocabulary." Diss., Virginia Tech, 2005. http://hdl.handle.net/10919/25955.
Full textAbstract:
This research focuses on the strategies that native English speakers use as they learn to speak and write Chinese vocabulary words in the first year of an elementary Chinese class. The main research question was: what strategies do native English-speaking beginning learners of Chinese use to learn Chinese vocabulary words in their speaking and writing? The study was conducted at a medium-sized comprehensive university in the Southeastern U.S. The study drew from concepts and theories in second language acquisition and psycholinguistic studies. A random sampling of four students was selected in their first year of Chinese study for qualitative analyses. Data were collected from demographic student surveys, reflection papers, interviews, observation and field notes, weekly diary of the students and Strategies Inventory for Language Learning (SILL).
The conclusions from this study provide insight as to how students of this demographic approach the challenge of learning Chinese. From this study, a clear picture emerges that students use different strategies to learn Chinese. Some students respond better to sound while others are more visually based learners. However, in this study, students used combinations of audio, visual, and kinesthetic learning techniques. The tonality of spoken Chinese was one of the most difficult skills to master and this aspect of the language frustrated many students. This is a widely recognized problem with Chinese education. Nevertheless, students enjoyed the artistic nature of Chinese characters and for the most part enjoyed writing them. This element can be emphasized in Chinese instruction to motivate students and appeal to visual learners. Similarly, integrating instruction on Chinese culture into language classes made the Elementary Chinese curriculum more appealing to students. Using native Chinese speakers from the local community in the language curriculum, reinforced classroom instruction, made the instruction more relevant, and increased student interest. Encouraging students to attend Chinese cultural events in the community had many of the same positive benefits for students. The motivations for learning revealed in this study are very interesting and support earlier studies of Chinese learners. Personal and profession interests as well as a combination of both these factors were the most commonly cited reasons for learning Chinese. Maintaining proper motivation is a pivotal factor that determines the success of many elementary learners including the students in this study. When students lost their motivation, interest in the curriculum and learning declined as well. Teachers need to be aware of motivations and attempt to foster them in individual students in order to maximize the learning experience.Ph. D.
12
EL-Manzalawy, Yasser. "Machine learning approaches for epitope prediction." [Ames, Iowa : Iowa State University], 2008.
Find full text
13
Marsden-Huggins, John. "Towards an understanding of ESL students' approaches to learning: a study of conceptions of learning, perceptions of situational demands, learning approaches and learning outcomes." Doctoral thesis, University of Cape Town, 1994. http://hdl.handle.net/11427/15993.
Full textAbstract:
Includes bibliographical references.An hypothesised relationship between levels of proficiency in English of ESL (English as a Second Language) students and the approaches to learning which they adopt, in situations in which English is the language of instruction, is the focus of this study. An attempt was made to identify the extent to which students, who are required to learn in a second language, adopt undesirable approaches to learning as a consequence of linguistic or cultural factors. Such students appear to adopt reproductive strategies to pass examinations and retain only isolated pieces of information for practical application. In a sense, they graduate but remain unqualified. Quantitative responses of 307 students, relating to their contextualised perceptions of the demands of the learning situation, were gathered and analysed using a learning approach categorisation procedure. Qualitative responses of 120 students, relating to their descriptions of the context and content of learning, were gathered in semi-structured interviews to supplement and enrich the quantitive data collected. Levels of proficiency in the language of instruction were measured using integrative tests of comprehension of spoken discourse and written texts presented in actual lecture situations. Students were given the opportunity to rate the lectures and reading material from which they were expected to learn and self-esteem was measured as a construct considered likely to affect perceptions of the demands of the learning situation. Concurrently with the above, a group of students from each of 3 year groups was taught a new topic over a short series of lectures and tested for understanding in the language of instruction. Balanced groups, from each of the 3 year groups, were taught the same topic and tested for understanding in the mother-tongue. This procedure was subsequently replicated with a second topic of similar complexity, across all three year groups, with languages switched. Critical aspects of the teaching/learning situation were kept constant. These procedures provided compelling evidence, after analysis of quantitative and qualitative data, of a relationship between proficiency in the language of instruction and the ways in which students engage in learning tasks. Difficulty with the language of instruction appears to increase the demands of the learning situation and the likelihood of adopting reproducing strategies, which are not normally associated with success in terms of learning outcomes.
14
Del, Valle Rodrigo. "Online learning learner characteristics and their approaches to managing learning /." [Bloomington, Ind.] : Indiana University, 2006. http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:3204535.
Full textAbstract:
Thesis (Ph.D.)--Indiana University, Dept. of Instructional Systems Technology of the School of Education, 2006.Source: Dissertation Abstracts International, Volume: 67-01, Section: A, page: 0152. Adviser: Thomas M. Duffy. "Title from dissertation home page (viewed Jan. 8, 2007)."
15
Katzenbach, Michael. "Individual Approaches in Rich Learning Situations Material-based Learning with Pinboards." Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2012. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-80328.
Full textAbstract:
Active Approaches provide chances for individual, comprehension-oriented learning and can facilitate the acquirement of general mathematical competencies. Using the example of pinboards, which were developed for different areas of the secondary level, workshop participants experience, discuss and further develop learning tasks, which can be used for free activities, for material based concept
formation, for coping with heterogeneity, for intelligent exercises, as tool for the presentation of students’ work and as basis for games. The material also allows some continuous movements and can thus prepare an insightful usage of dynamic geometry programs. Central Part of the workshop is a work-sharing group work with learning tasks for grades 5 to 8. The workshop will close with a discussion of general aspects of material-based learning.
16
Dalton-Brits, E., and M. Viljoen. "Personality traits and learning approaches : are they influencing the learning process?" Journal for New Generation Sciences, Vol 8, Issue 3: Central University of Technology, Free State, Bloemfontein, 2010. http://hdl.handle.net/11462/565.
Full textAbstract:
Published ArticleThe relationship between the big five personality traits, Extraversion, Agreeableness Neuroticism, Conscientiousness and Openness to Experience and deep and surface approaches to learning forms the basis of this article. The findings of a research study in this milieu will be presented to prove that earlier studies in this field have been upheld, but that an important deviation has occurred on certain levels of personality. A students way of learning implies the type of learning that is taking place. Ultimately we as lecturers want to encourage deep learning as this stimulates retention of information, important in production of students that are ready for employment.
17
Pon, Kumar Steven Spielberg. "Deep reinforcement learning approaches for process control." Thesis, University of British Columbia, 2017. http://hdl.handle.net/2429/63810.
Full textAbstract:
The conventional and optimization based controllers have been used in process industries for more than two decades. The application of such controllers on complex systems could be computationally demanding and may require estimation of hidden states. They also require constant tuning, development of a mathematical model (first principle or empirical), design of control law which are tedious. Moreover, they are not adaptive in nature. On the other hand, in the recent years, there has been significant progress in the fields of computer vision and natural language processing that followed the success of deep learning. Human level control has been attained in games and physical tasks by combining deep learning with reinforcement learning. They were also able to learn the complex go game which has states more than number of atoms in the universe. Self-Driving cars, machine translation, speech recognition etc started to gain advantage of these powerful models. The approach to all of them involved problem formulation as a learning problem. Inspired by these applications, in this work we have posed process control problem as a learning problem to build controllers to address the limitations existing in current controllers.Applied Science, Faculty of
Chemical and Biological Engineering, Department of
Graduate
18
Stamp, D. I. "Machine learning approaches to complex time series." Thesis, University of Liverpool, 1999. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.399317.
Full textAbstract:
It has been noted that there are numerous similarities between the behaviour of chaotic and stochastic systems. The theoretical links between chaotic and stochastic systems are investigated based on the evolution of the density of dynamics and an equivalency relationship based on the invariant measure of an ergodic system. It is shown that for simple chaotic systems an equivalent stochastic model can be analytically derived when the initial position in state space is only known to a limited precision. Based on this a new methodology for the modelling of complex nonlinear time series displaying chaotic behaviour with stochastic models is proposed. This consists of using a stochastic model to learn the evolution of the density of the dynamics of the chaotic system by estimating initial and transitional density functions directly from a time series. A number of models utilising this methodology are proposed, based on Markov chains and hidden Markov models. These are implemented and their performance and characteristics compared using computer simulation with several standard techniques.
19
Kwan, Sze-wai David, and 關思偉. "Thinking styles, learning approaches, and academic achievement." Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2000. http://hub.hku.hk/bib/B31961666.
Full text
20
Veropoulos, Konstantinos. "Machine learning approaches to medical decision making." Thesis, University of Bristol, 2001. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.367661.
Full text
21
Mbipom, Blessing. "Knowledge driven approaches to e-learning recommendation." Thesis, Robert Gordon University, 2018. http://hdl.handle.net/10059/3121.
Full textAbstract:
Learners often have difficulty finding and retrieving relevant learning materials to support their learning goals because of two main challenges. The vocabulary learners use to describe their goals is different from that used by domain experts in teaching materials. This challenge causes a semantic gap. Learners lack sufficient knowledge about the domain they are trying to learn about, so are unable to assemble effective keywords that identify what they wish to learn. This problem presents an intent gap. The work presented in this thesis focuses on addressing the semantic and intent gaps that learners face during an e-Learning recommendation task. The semantic gap is addressed by introducing a method that automatically creates background knowledge in the form of a set of rich learning-focused concepts related to the selected learning domain. The knowledge of teaching experts contained in e-Books is used as a guide to identify important domain concepts. The concepts represent important topics that learners should be interested in. An approach is developed which leverages the concept vocabulary for representing learning materials and this influences retrieval during the recommendation of new learning materials. The effectiveness of our approach is evaluated on a dataset of Machine Learning and Data Mining papers, and our approach outperforms benchmark methods. The results confirm that incorporating background knowledge into the representation of learning materials provides a shared vocabulary for experts and learners, and this enables the recommendation of relevant materials. We address the intent gap by developing an approach which leverages the background knowledge to identify important learning concepts that are employed for refining learners' queries. This approach enables us to automatically identify concepts that are similar to queries, and take advantage of distinctive concept terms for refining learners' queries. Using the refined query allows the search to focus on documents that contain topics which are relevant to the learner. An e-Learning recommender system is developed to evaluate the success of our approach using a collection of learner queries and a dataset of Machine Learning and Data Mining learning materials. Users with different levels of expertise are employed for the evaluation. Results from experts, competent users and beginners all showed that using our method produced documents that were consistently more relevant to learners than when the standard method was used. The results show the benefits in using our knowledge driven approaches to help learners find relevant learning materials.
22
McClellan, Timothy. "Creative learning approaches for undergraduate self-development." Thesis, University of Southampton, 2013. https://eprints.soton.ac.uk/368989/.
Full textAbstract:
This thesis investigates creativity in the undergraduate curriculum and how students respond to creative approaches to learning within their studies. Specifically, the thesis considers how the use of multiple creative learning methods may enhance undergraduate learning and the role that creative visualisation and guided imagery can play in this experience. The thesis presents the learning stories of six undergraduates in the main study who took one of these modules. Interviews were conducted and a range of other documentary data, such as learning journals and assignments, was collected and analysed in order to detail each student’s journey through and experience of the module. The analysis is presented in three separate sections; firstly, as individual student case studies; secondly, as a thematic cross-case analysis; and thirdly, as a synthesis of the data with theoretical constructs and current debates surrounding creativity in higher education with conclusions and recommendations for individual and sector practice. The thesis discusses the ‘messy’ nature of research, highlights the compromises and difficulties inherent in a PhD project and illustrates how these issues were overcome. The work also reflects on the researcher’s own PhD learning journey and identifies a number of themes that influence the efficacy of the teaching of creative skills in undergraduate programmes. The thesis proposes a number of new models that have been integrated into the author’s own teaching and that have wider implications for the teaching of transferable skills in creativity and creative thinking in higher education for practice-based and non-vocational programmes as well as consultancy opportunities for industry. New knowledge proposed within the thesis includes a refined model of student engagement and a model to plot the student journey of self-discovery. The thesis also offers a critique of and guidelines for the use of guided imagery to promote student creativity in higher education.
23
Dang, Ha Xuan. "Mold Allergomics: Comparative and Machine Learning Approaches." Diss., Virginia Tech, 2014. http://hdl.handle.net/10919/64205.
Full textAbstract:
Fungi are one of the major organisms that cause allergic disease in human. A number of proteins from fungi have been found to be allergenic or possess immunostimulatory properties. Identifying and characterizing allergens from fungal genomes will help facilitate our understanding of the mechanism underlying host-pathogen interactions in allergic diseases. Currently, there is a lack of tools that allow us to rapidly and accurately predict allergens from whole genomes. In the context of whole genome annotation, allergens are rare compared to non-allergens and thus the data is considered highly skewed. In order to achieve a confident set of predicted allergens from a genome, false positive rates must be lowered. Current allergen prediction tools often produce many false positives when applied to large-scale data set such as whole genomes, and thus lower the precision. Moreover, the most accurate tools are relatively slow because they use sequence alignment to construct feature vectors for allergen classifiers. This dissertation presents computational approaches in characterizing the allergen repertoire in fungal genomes as part of the whole genome studies of Alternaria, an important allergenic/opportunistic human pathogenic fungus and necrotrophic plant parasite. In these studies, the genomes of multiple Alternaria species were characterized for the first time. Functional elements (e.g. genes, proteins) were first identified and annotated from these genomes using computational tools. Protein annotation and comparative genomics approaches revealed the link between Alternaria genotypes and its prolific saprophytic lifestyle that provides at least a partial explanation for the development of pathological relationships between Alternaria and humans. A machine learning based tool (Allerdictor) was developed to address the neglected problem of allergen prediction in highly skewed large-scale data sets. Allerdictor exhibited high precision over high recall at fast speed and thus it is a more practical tool for large-scale allergen annotation compared with existing tools. Allerdictor was then used together with a comparative genomics approach to survey the allergen repertoire of known allergenic fungi. We predicted a number of mold allergens that have not been experimentally characterized. These predicted allergens are potential candidates for further experimental and clinical validation. Our approaches will not only facilitate the study of allergens in the increasing number of sequenced fungal genomes but also will be useful for allergen annotation in other species and rapid prescreening of synthesized sequences for potential allergens.Ph. D.
24
Sathyan, Anoop. "Intelligent Machine Learning Approaches for Aerospace Applications." University of Cincinnati / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=ucin1491558309625214.
Full text
25
Tortajada, Velert Salvador. "Incremental Learning approaches to Biomedical decision problems." Doctoral thesis, Universitat Politècnica de València, 2012. http://hdl.handle.net/10251/17195.
Full textAbstract:
During the last decade, a new trend in medicine is transforming the nature of healthcare from reactive to proactive. This new paradigm is changing into a personalized medicine where the prevention, diagnosis, and treatment of disease is focused on individual patients. This paradigm is known as P4 medicine. Among other key benefits, P4 medicine aspires to detect diseases at an early stage and introduce diagnosis to stratify patients and diseases to select the optimal therapy based on individual observations and taking into account the patient outcomes to empower the physician, the patient, and their communication.
This paradigm transformation relies on the availability of complex multi-level biomedical data that are increasingly accurate, since it is possible to find exactly the needed information, but also exponentially noisy, since the access to that information is more and more challenging. In order to take advantage of this information, an important effort is being made in the last decades to digitalize medical records and to develop new mathematical and computational methods for extracting maximum knowledge from patient records, building dynamic and disease-predictive models from massive amounts of integrated clinical and biomedical data. This requirement enables the use of computer-assisted Clinical Decision Support Systems for the management of individual patients.
The Clinical Decision Support System (CDSS) are computational systems that provide precise and specific knowledge for the medical decisions to be adopted for diagnosis, prognosis, treatment and management of patients. The CDSS are highly related to the concept of evidence-based medicine since they infer medical knowledge from the biomedical databases and the acquisition protocols that are used for the development of the systems, give computational support based on evidence for the clinical practice, and evaluate the performance and the added value of the solution for each specific medical problem.Tortajada Velert, S. (2012). Incremental Learning approaches to Biomedical decision problems [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/17195
Palancia
26
Wang, Yiqing. "Two Bayesian learning approaches to image processing." Thesis, Cachan, Ecole normale supérieure, 2015. http://www.theses.fr/2015DENS0007/document.
Full textAbstract:
Cette thèse porte sur deux méthodes à patch en traitement d’images dans le cadre de minimisation du risque Bayésien. Nous décrivons un mélange d’analyses factorielles pour modéliser la loi à priori des patchs dans une seule image et l’appliquons au débruitage et à l’inpainting. Nous étudions aussi les réseaux de neurones à multi-couches d’un point de vue probabiliste comme un outil permettant d’approcher l’espérance conditionnelle, ce qui ouvre quelques voies pour réduire leurs tailles et coût d’apprentissageThis work looks at two patch based image processing methods in a Bayesian risk minimization framework. We describe a Gaussian mixture of factor analyzers for local prior modelling and apply it in the context of image denoising and inpainting. We also study multilayer neural networks from a probabilistic perspective as a tool for conditional expectation approximation, which suggests ways to reduce their sizes and training cost
27
Tuli, Sabrina Hoque. "Small Face Detection with Deep Learning Approaches." Thesis, Curtin University, 2021. http://hdl.handle.net/20.500.11937/86208.
Full textAbstract:
This thesis considers small face detection in uncontrolled environments and develops robust deep learning approaches for this challenging problem. A novel multi-scale face detector is developed by integrating novel anchor design, efficient regression loss and additional detection layers. Several multi-scale dense convolutional networks are developed to boost up the detection of small faces. Experimental results on public face databases demonstrate that the proposed methods outperform the state-of-the-art methods (e.g. YOLOv3) for detection of small faces.
28
Taheri, Sona. "Learning Bayesian networks based on optimization approaches." Thesis, University of Ballarat, 2012. http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/36051.
Full textAbstract:
Learning accurate classifiers from preclassified data is a very active research topic in machine learning and artifcial intelligence. There are numerous classifier paradigms, among which Bayesian Networks are very effective and well known in domains with uncertainty. Bayesian Networks are widely used representation frameworks for reasoning with probabilistic information. These models use graphs to capture dependence and independence relationships between feature variables, allowing a concise representation of the knowledge as well as efficient graph based query processing algorithms. This representation is defined by two components: structure learning and parameter learning. The structure of this model represents a directed acyclic graph. The nodes in the graph correspond to the feature variables in the domain, and the arcs (edges) show the causal relationships between feature variables. A directed edge relates the variables so that the variable corresponding to the terminal node (child) will be conditioned on the variable corresponding to the initial node (parent). The parameter learning represents probabilities and conditional probabilities based on prior information or past experience. The set of probabilities are represented in the conditional probability table. Once the network structure is constructed, the probabilistic inferences are readily calculated, and can be performed to predict the outcome of some variables based on the observations of others. However, the problem of structure learning is a complex problem since the number of candidate structures grows exponentially when the number of feature variables increases. This thesis is devoted to the development of learning structures and parameters in Bayesian Networks. Different models based on optimization techniques are introduced to construct an optimal structure of a Bayesian Network. These models also consider the improvement of the Naive Bayes' structure by developing new algorithms to alleviate the independence assumptions. We present various models to learn parameters of Bayesian Networks; in particular we propose optimization models for the Naive Bayes and the Tree Augmented Naive Bayes by considering different objective functions. To solve corresponding optimization problems in Bayesian Networks, we develop new optimization algorithms. Local optimization methods are introduced based on the combination of the gradient and Newton methods. It is proved that the proposed methods are globally convergent and have superlinear convergence rates. As a global search we use the global optimization method, AGOP, implemented in the open software library GANSO. We apply the proposed local methods in the combination with AGOP. Therefore, the main contributions of this thesis include (a) new algorithms for learning an optimal structure of a Bayesian Network; (b) new models for learning the parameters of Bayesian Networks with the given structures; and finally (c) new optimization algorithms for optimizing the proposed models in (a) and (b). To validate the proposed methods, we conduct experiments across a number of real world problems. Print version is available at: http://library.federation.edu.au/record=b1804607~S4Doctor of Philosophy
29
Amuru, SaiDhiraj. "Intelligent Approaches for Communication Denial." Diss., Virginia Tech, 2015. http://hdl.handle.net/10919/56695.
Full textAbstract:
Spectrum supremacy is a vital part of security in the modern era. In the past 50 years, a great deal of work has been devoted to designing defenses against attacks from malicious nodes (e.g., anti-jamming), while significantly less work has been devoted to the equally important task of designing effective strategies for denying communication between enemy nodes/radios within an area (e.g., jamming). Such denial techniques are especially useful in military applications and intrusion detection systems where untrusted communication must be stopped. In this dissertation, we study these offensive attack procedures, collectively termed as communication denial. The communication denial strategies studied in this dissertation are not only useful in undermining the communication between enemy nodes, but also help in analyzing the vulnerabilities of existing systems.
A majority of the works which address communication denial assume that knowledge about the enemy nodes is available a priori. However, recent advances in communication systems creates the potential for dynamic environmental conditions where it is difficult and most likely not even possible to obtain a priori information regarding the environment and the nodes that are present in it. Therefore, it is necessary to have cognitive capabilities that enable the attacker to learn the environment and prevent enemy nodes from accessing valuable spectrum, thereby denying communication.
In this regard, we ask the following question in this dissertation ``Can an intelligent attacker learn and adapt to unknown environments in an electronic warfare-type scenario?" Fundamentally speaking, we explore whether existing machine learning techniques can be used to address such cognitive scenarios and, if not, what are the missing pieces that will enable an attacker to achieve spectrum supremacy by denying an enemy the ability to communicate? The first task in achieving spectrum supremacy is to identify the signal of interest before it can be attacked. Thus, we first address signal identification, specifically modulation classification, in practical wireless environments where the interference is often non-Gaussian. Upon identifying the signal of interest, the next step is to effectively attack the victim signals in order to deny communication. We present a rigorous fundamental analysis regarding the attackers performance, in terms of achieving communication denial, in practical communication settings. Furthermore, we develop intelligent approaches for communication denial that employ novel machine learning techniques to attack the victim either at the physical layer, the MAC layer, or the network layer. We rigorously investigate whether or not these learning techniques enable the attacker to approach the fundamental performance limits achievable when an attacker has complete knowledge of the environment. As a result of our work, we debunk several myths about communication denial strategies that were believed to be true mainly because incorrect system models were previously considered and thus the wrong questions were answered.Ph. D.
30
Strang, Alison Bridget. "A model of learning : an investigation of technicians' approaches to open learning." Thesis, University College London (University of London), 1990. http://discovery.ucl.ac.uk/10018494/.
Full textAbstract:
The work arose from an applied research project commissioned by the Manpower Services Commission. The aim of the project was to produce guidelines for effective learning in the Open Tech, based on a thoroughly researched understanding of open learning at technician level. A review of the literature led to the proposal that the research should examine technicians' experiences of learning from a phenomenological perspective, with particular attention to the intentions and meanings underlying their approaches. In response to this proposal, a metatheory was formulated to establish the assumptions on which the research should be based. The metatheory incorporated a view of man as a natural learner, implying that the research should seek to understand why natural learning behaviour is inhibited. The view of science specified that the outcome of the research should be the development of an appropriate and useful model of technician open learning. The development of this model was 'grounded' in the empirical study of British Telecom open learning students. The study incorporated both an experimental learning, task and focused interviewing. Notable dimensions emerging from learners' accounts of their experiences of learning included: orientations to study, conceptions of learning, and locus of control in learning. Relationships between these dimensions were explored and a series of 'procedural steps' was proposed, which outlines the key processes necessary to effective learning in this context. This empirical analysis led to the formulation of the 'multi-dimensional' model of learning, which suggests that meaningful learning arises as a learner interacts with a task in pursuit of his own learning intentions. On the basis of this definition it was possible to identify the qualities of meaningful learning, and to recognise the equivalence of the notions of meaningful, effective and autonomous learning. The model was tested and elaborated, using data from a further study of a different group of technicians undertaking open learning courses. Finally, the practical applications of the model for the Open Tech were explored.
31
Ollerenshaw, Alison. "Learning through multimedia : the roles of prior knowledge and approaches to learning." Thesis, The Author [Mt.Helen, Vic.] :, 1999. http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/44369.
Full textAbstract:
The effects of text-supplementing illustrations have been generally well etablished (Mayer, Bove, Bryman, Mars & Tapangco, 1996). However, these effects are not universal, and are influenced by learner factors including student approaches to learning and prior knowldge (Ollerenshaw, Aidman & Kidd, 1997)....Master of Applied Science (Psychology)
32
Lai, Ling-yan Edith. "Effects of cooperative learning on student learning outcomes and approaches to learning in sixth form geography." Click to view the E-thesis via HKUTO, 1991. http://sunzi.lib.hku.hk/HKUTO/record/B38627292.
Full text
33
Stanzione, Vincenzo Maria. "Developing a new approach for machine learning explainability combining local and global model-agnostic approaches." Master's thesis, Alma Mater Studiorum - Università di Bologna, 2022. http://amslaurea.unibo.it/25480/.
Full textAbstract:
The last couple of past decades have seen a new flourishing season for the Artificial
Intelligence, in particular for Machine Learning (ML). This
is reflected in the great number of fields that are employing ML solutions to overcome a broad spectrum of problems.
However, most of the last employed ML models have a black-box
behavior. This means that given a certain input, we are not able to understand
why one of these models produced a certain output or made a certain
decision. Most of the time, we are not interested in knowing what
and how the model is thinking, but if we think of a model which makes extremely critical
decisions or takes
decisions that have a heavy result on people’s lives, in these cases explainability is a duty.
A great variety of techniques to perform global or local explanations are available. One of the most widespread is Local Interpretable Model-Agnostic
Explanations (LIME), which creates a local linear model in the proximity of an
input to understand in which way each feature contributes to the final output.
However, LIME is not immune from instability problems and sometimes to
incoherent predictions. Furthermore, as a local explainability technique, LIME
needs to be performed for each different input that we want to explain.
In this work, we have been inspired by the LIME approach for linear models
to craft a novel technique. In combination with the Model-based Recursive Partitioning
(MOB), a brand-new score function to assess the quality of a partition
and the usage of Sobol quasi-Montecarlo sampling, we developed a new global
model-agnostic explainability technique we called Global-Lime.
Global-Lime
is capable of giving a global understanding of the original ML
model, through an ensemble of spatially not overlapped hyperplanes, plus a local
explanation for a certain output considering only the corresponding linear
approximation. The idea is to train the black-box model and then supply along
with it its explainable version.
34
Tweed, Roger Gordon. "Learning considered within a cultural context Confucian and Socratic approaches /." online access from Digital Dissertation Consortium access full-text, 2000. http://libweb.cityu.edu.hk/cgi-bin/er/db/ddcdiss.pl?NQ56637.
Full text
35
Khan, Umair. "Self-supervised deep learning approaches to speaker recognition." Doctoral thesis, Universitat Politècnica de Catalunya, 2021. http://hdl.handle.net/10803/671496.
Full textAbstract:
In speaker recognition, i-vectors have been the state-of-the-art unsupervised technique over the last few years, whereas x-vectors is becoming the state-of-the-art supervised technique, these days. Recent advances in Deep Learning (DL) approaches to speaker recognition have improved the performance but are constrained to the need of labels for the background data. In practice, labeled background data is not easily accessible, especially when large training data is required. In i-vector based speaker recognition, cosine and Probabilistic Linear Discriminant Analysis (PLDA) are the two basic scoring techniques. Cosine scoring is unsupervised whereas PLDA parameters are typically trained using speaker-labeled background data. This makes a big performance gap between these two scoring techniques. The question is: how to fill this performance gap without using speaker labels for the background data? In this thesis, the above mentioned problem has been addressed using DL approaches without using and/or limiting the use of labeled background data. Three DL based proposals have been made.
In the first proposal, a Restricted Boltzmann Machine (RBM) vector representation of speech is proposed for the tasks of speaker clustering and tracking in TV broadcast shows. This representation is referred to as RBM vector. The experiments on AGORA database show that in speaker clustering the RBM vectors gain a relative improvement of 12% in terms of Equal Impurity (EI). For speaker tracking task RBM vectors are used only in the speaker identification part, where the relative improvement in terms of Equal Error Rate (EER) is 11% and 7% using cosine and PLDA scoring, respectively.
In the second proposal, DL approaches are proposed in order to increase the discriminative power of i-vectors in speaker verification. We have proposed the use of autoencoder in several ways. Firstly, an autoencoder will be used as a pre-training for a Deep Neural Network (DNN) using a large amount of unlabeled background data. Then, a DNN classifier will be trained using relatively small labeled data. Secondly, an autoencoder will be trained to transform i-vectors into a new representation to increase their discriminative power. The training will be carried out based on the nearest neighbor i-vectors which will be chosen in an unsupervised manner. The evaluation was performed on VoxCeleb-1 database. The results show that using the first system, we gain a relative improvement of 21% in terms of EER, over i-vector/PLDA. Whereas, using the second system, a relative improvement of 42% is gained. If we use the background data in the testing part, a relative improvement of 53% is gained.
In the third proposal, we will train a self-supervised end-to-end speaker verification system. The idea is to utilize impostor samples along with the nearest neighbor samples to make client/impostor pairs in an unsupervised manner. The architecture will be based on a Convolutional Neural Network (CNN) encoder, trained as a siamese network with two branch networks. Another network with three branches will also be trained using triplet loss, in order to extract unsupervised speaker embeddings. The experimental results show that both the end-to-end system and the speaker embeddings, despite being unsupervised, show a comparable performance to the supervised baseline. Moreover, their score combination can further improve the performance.
The proposed approaches for speaker verification have respective pros and cons. The best result was obtained using the nearest neighbor autoencoder with a disadvantage of relying on background i-vectors in the testing. On the contrary, the autoencoder pre-training for DNN is not bound by this factor but is a semi-supervised approach. The third proposal is free from both these constraints and performs pretty reasonably. It is a self-supervised approach and it does not require the background i-vectors in the testing phase.Los avances recientes en Deep Learning (DL) para el reconocimiento del hablante están mejorado el rendimiento de los sistemas tradicionales basados en i-vectors. En el reconocimiento de locutor basado en i-vectors, la distancia coseno y el análisis discriminante lineal probabilístico (PLDA) son las dos técnicas más usadas de puntuación. La primera no es supervisada, pero la segunda necesita datos etiquetados por el hablante, que no son siempre fácilmente accesibles en la práctica. Esto crea una gran brecha de rendimiento entre estas dos técnicas de puntuación. La pregunta es: ¿cómo llenar esta brecha de rendimiento sin usar etiquetas del hablante en los datos de background? En esta tesis, el problema anterior se ha abordado utilizando técnicas de DL sin utilizar y/o limitar el uso de datos etiquetados. Se han realizado tres propuestas basadas en DL. En la primera, se propone una representación vectorial de voz basada en la máquina de Boltzmann restringida (RBM) para las tareas de agrupación de hablantes y seguimiento de hablantes en programas de televisión. Los experimentos en la base de datos AGORA, muestran que en agrupación de hablantes los vectores RBM suponen una mejora relativa del 12%. Y, por otro lado, en seguimiento del hablante, los vectores RBM,utilizados solo en la etapa de identificación del hablante, muestran una mejora relativa del 11% (coseno) y 7% (PLDA). En la segunda, se utiliza DL para aumentar el poder discriminativo de los i-vectors en la verificación del hablante. Se ha propuesto el uso del autocodificador de varias formas. En primer lugar, se utiliza un autocodificador como preentrenamiento de una red neuronal profunda (DNN) utilizando una gran cantidad de datos de background sin etiquetar, para posteriormente entrenar un clasificador DNN utilizando un conjunto reducido de datos etiquetados. En segundo lugar, se entrena un autocodificador para transformar i-vectors en una nueva representación para aumentar el poder discriminativo de los i-vectors. El entrenamiento se lleva a cabo en base a los i-vectors vecinos más cercanos, que se eligen de forma no supervisada. La evaluación se ha realizado con la base de datos VoxCeleb-1. Los resultados muestran que usando el primer sistema obtenemos una mejora relativa del 21% sobre i-vectors, mientras que usando el segundo sistema, se obtiene una mejora relativa del 42%. Además, si utilizamos los datos de background en la etapa de prueba, se obtiene una mejora relativa del 53%. En la tercera, entrenamos un sistema auto-supervisado de verificación de locutor de principio a fin. Utilizamos impostores junto con los vecinos más cercanos para formar pares cliente/impostor sin supervisión. La arquitectura se basa en un codificador de red neuronal convolucional (CNN) que se entrena como una red siamesa con dos ramas. Además, se entrena otra red con tres ramas utilizando la función de pérdida triplete para extraer embeddings de locutores. Los resultados muestran que tanto el sistema de principio a fin como los embeddings de locutores, a pesar de no estar supervisados, tienen un rendimiento comparable a una referencia supervisada. Cada uno de los enfoques propuestos tienen sus pros y sus contras. El mejor resultado se obtuvo utilizando el autocodificador con el vecino más cercano, con la desventaja de que necesita los i-vectors de background en el test. El uso del preentrenamiento del autocodificador para DNN no tiene este problema, pero es un enfoque semi-supervisado, es decir, requiere etiquetas de hablantes solo de una parte pequeña de los datos de background. La tercera propuesta no tienes estas dos limitaciones y funciona de manera razonable. Es un en
36
Fasel, Ian Robert. "Learning real-time object detectors probabilistic generative approaches /." Connect to a 24 p. preview or request complete full text in PDF format. Access restricted to UC campuses, 2006. http://wwwlib.umi.com/cr/ucsd/fullcit?p3216357.
Full textAbstract:
Thesis (Ph. D.)--University of California, San Diego, 2006.Title from first page of PDF file (viewed July 24, 2006). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 87-91).
37
Gulcehre, Caglar. "Two Approaches For Collective Learning With Language Games." Master's thesis, METU, 2011. http://etd.lib.metu.edu.tr/upload/12613109/index.pdf.
Full textAbstract:
Recent studies in cognitive science indicate that language has an important social function. The structure and knowledge of language emerges from the processes of human communication together with the domain-general cognitive processes. Each individual of a community interacts socially with a limited number of peers. Nevertheless societies are characterized by their stunning global regularities. By dealing with the language as a complex adaptive system, we are able to analyze how languages change and evolve over time. Multi-agent computational simulations assist scientists from different disciplines to build several language emergence scenarios. In this thesis several simulations are implemented and tested in order to categorize examples in a test data set efficiently and accurately by using a population of agents interacting by playing categorization games inspired by L. Steels's naming game. The emergence of categories throughout interactions between a population of agents in the categorization games are analyzed. The test results of categorization games as a model combination algorithm with various machine learning algorithms on different data sets have shown that categorization games can have a comparable performance with fast convergence.
38
Nori, Nozomi. "Machine Learning Approaches for Personalized Clinical Risk Modeling." 京都大学 (Kyoto University), 2017. http://hdl.handle.net/2433/225729.
Full text
39
Zhang, Yue. "Discriminative learning approaches for statistical processing of Chinese." Thesis, University of Oxford, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.510405.
Full text
40
Trifonova, Neda. "Machine-learning approaches for modelling fish population dynamics." Thesis, Brunel University, 2016. http://bura.brunel.ac.uk/handle/2438/13386.
Full textAbstract:
Ecosystems consist of complex dynamic interactions among species and the environment, the understanding of which has implications for predicting the environmental response to changes in climate and biodiversity. Understanding the nature of functional relationships (such as prey-predator) between species is important for building predictive models. However, modelling the interactions with external stressors over time and space is also essential for ecosystem-based approaches to fisheries management. With the recent adoption of more explorative tools, like Bayesian networks, in predictive ecology, fewer assumptions can be made about the data and complex, spatially varying interactions can be recovered from collected field data and combined with existing knowledge. In this thesis, we explore Bayesian network modelling approaches, accounting for latent effects to reveal species dynamics within geographically different marine ecosystems. First, we introduce the concept of functional equivalence between different fish species and generalise trophic structure from different marine ecosystems in order to predict influence from natural and anthropogenic sources. The importance of a hidden variable in fish community change studies of this nature was acknowledged because it allows causes of change which are not purely found within the constrained model structure. Then, a functional network modelling approach was developed for the region of North Sea that takes into consideration unmeasured latent effects and spatial autocorrelation to model species interactions and associations with external factors such as climate and fisheries exploitation. The proposed model was able to produce novel insights on the ecosystem's dynamics and ecological interactions mainly because it accounts for the heterogeneous nature of the driving factors within spatially differentiated areas and their changes over time. Finally, a modified version of this dynamic Bayesian network model was used to predict the response of different ecosystem components to change in anthropogenic and environmental factors. Through the development of fisheries catch, temperature and productivity scenarios, we explore the future of different fish and zooplankton species and examine what trends of fisheries exploitation and environmental change are potentially beneficial in terms of ecological stability and resilience. Thus, we were able to provide a new data-driven modelling approach which might be beneficial to give strategic advice on potential response of the system to pressure.
41
Muppala, Sireesha. "Multi-tier Internet service management| Statistical learning approaches." Thesis, University of Colorado at Colorado Springs, 2013. http://pqdtopen.proquest.com/#viewpdf?dispub=3560749.
Full textAbstract:
Modern Internet services are multi-tiered and are typically hosted in virtualized shared platforms. While facilitating flexible service deployment, multi-tier architecture introduces significant challenges for Quality of Service (QoS) provisioning in hosted Internet services. Complex inter-tier dependencies and dynamic bottleneck tier shift are challenges inherent to tiered architectures. Hard-to-predict and bursty session-based Internet workloads further magnify this complexity. Virtualization of shared platforms adds yet another layer of complication in managing the hosted multi-tier Internet services.
We consider three critical aspects of Internet service management for improved performance and quality of service provisioning : admission control, dynamic resource provisioning and service differentiation. This thesis concentrates on statistical learning based approaches for multi-tier Internet service management to achieve efficient, balanced and scalable services. Statistical learning techniques are capable of solving complex dynamic problems through learning and adaptation with no priori domain-specific knowledge. We explore the effectiveness of supervised and unsupervised learning in managing multi-tier Internet services.
First, we develop a session based admission control strategy to improve session throughput of multi- tier Internet services. Using a supervised bayesian network, it achieves coordination among multiple tiers resulting in a balanced service. Second, we promote session-slowdown, a novel session-oriented metric for user perceived performance. We develop a regression based dynamic resource provisioning strategy, which utilizes a combination of offline training and online monitoring, for session slowdown guarantees in multi-tier systems. Third, we develop a reinforcement learning based coordinated combination of admission control and adaptive resource management for multi-tier Internet service differentiation and performance improvement in a shared virtualized platform. It addresses limitations of supervised learning by integrating model-independence of reinforcement learning and self-learning of neural networks for system scalability and agility. Finally, we develop an user interface based Monitoring and Management Console, intended for an administrator to monitor and fine tune the performance of hosted multi-tier Internet services.
We evaluate the developed management approaches using an e-commerce simulator and an implementation testbed on a virtualized blade server system hosting multi-tier RUBiS benchmark applications. Results demonstrate the effectiveness and efficiency of statistical learning approaches for QoS provisioning and performance improvement in virtualized multi-tier Internet services.
42
Li, Jerry Zheng. "Principled approaches to robust machine learning and beyond." Thesis, Massachusetts Institute of Technology, 2018. http://hdl.handle.net/1721.1/120382.
Full textAbstract:
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 305-320).
As we apply machine learning to more and more important tasks, it becomes increasingly important that these algorithms are robust to systematic, or worse, malicious, noise. Despite considerable interest, no efficient algorithms were known to be robust to such noise in high dimensional settings for some of the most fundamental statistical tasks for over sixty years of research. In this thesis we devise two novel, but similarly inspired, algorithmic paradigms for estimation in high dimensions in the presence of a small number of adversarially added data points. Both algorithms are the first efficient algorithms which achieve (nearly) optimal error bounds for a number fundamental statistical tasks such as mean estimation and covariance estimation. The goal of this thesis is to present these two frameworks in a clean and unified manner. We show that these insights also have applications for other problems in learning theory. Specifically, we show that these algorithms can be combined with the powerful Sum-of-Squares hierarchy to yield improvements for clustering high dimensional Gaussian mixture models, the first such improvement in over fifteen years of research. Going full circle, we show that Sum-of-Squares also can be used to improve error rates for robust mean estimation. Not only are these algorithms of interest theoretically, but we demonstrate empirically that we can use these insights in practice to uncover patterns in high dimensional data that were previously masked by noise. Based on our algorithms, we give new implementations for robust PCA, new defenses for data poisoning attacks for stochastic optimization, and new defenses for watermarking attacks on deep nets. In all of these tasks, we demonstrate on both synthetic and real data sets that our performance is substantially better than the state-of-the-art, often able to detect most to all corruptions when previous methods could not reliably detect any.
by Jerry Zheng Li.
Ph. D.
43
Kim, Hyun Soo M. Eng Massachusetts Institute of Technology. "Two new approaches for learning Hidden Markov Models." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/61287.
Full textAbstract:
Thesis (M. Eng.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2010.Cataloged from PDF version of thesis.
Includes bibliographical references (p. 99-100).
Hidden Markov Models (HMMs) are ubiquitously used in applications such as speech recognition and gene prediction that involve inferring latent variables given observations. For the past few decades, the predominant technique used to infer these hidden variables has been the Baum-Welch algorithm. This thesis utilizes insights from two related fields. The first insight is from Angluin's seminal paper on learning regular sets from queries and counterexamples, which produces a simple and intuitive algorithm that efficiently learns deterministic finite automata. The second insight follows from a careful analysis of the representation of HMMs as matrices and realizing that matrices hold deeper meaning than simply entities used to represent the HMMs. This thesis takes Angluin's approach and nonnegative matrix factorization and applies them to learning HMMs. Angluin's approach fails and the reasons are discussed. The matrix factorization approach is successful, allowing us to produce a novel method of learning HMMs. The new method is combined with Baum-Welch into a hybrid algorithm. We evaluate the algorithm by comparing its performance in learning selected HMMs to the Baum-Welch algorithm. We empirically show that our algorithm is able to perform better than the Baum-Welch algorithm for HMMs with at most six states that have dense output and transition matrices. For these HMMs, our algorithm is shown to perform 22.65% better on average by the Kullback-Liebler measure.
by Hyun Soo Kim.
M.Eng.
44
Potapov, Danila. "Supervised Learning Approaches for Automatic Structuring of Videos." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAM023/document.
Full textAbstract:
L'Interprétation automatique de vidéos est un horizon qui demeure difficile a atteindre en utilisant les approches actuelles de vision par ordinateur. Une des principales difficultés est d'aller au-delà des descripteurs visuels actuels (de même que pour les autres modalités, audio, textuelle, etc) pour pouvoir mettre en oeuvre des algorithmes qui permettraient de reconnaitre automatiquement des sections de vidéos, potentiellement longues, dont le contenu appartient à une certaine catégorie définie de manière sémantique. Un exemple d'une telle section de vidéo serait une séquence ou une personne serait en train de pêcher; un autre exemple serait une dispute entre le héros et le méchant dans un film d'action hollywoodien. Dans ce manuscrit, nous présentons plusieurs contributions qui vont dans le sens de cet objectif ambitieux, en nous concentrant sur trois tâches d'analyse de vidéos: le résumé automatique, la classification, la localisation temporelle.Tout d'abord, nous introduisons une approche pour le résumé automatique de vidéos, qui fournit un résumé de courte durée et informatif de vidéos pouvant être très longues, résumé qui est de plus adapté à la catégorie de vidéos considérée. Nous introduisons également une nouvelle base de vidéos pour l'évaluation de méthodes de résumé automatique, appelé MED-Summaries, ou chaque plan est annoté avec un score d'importance, ainsi qu'un ensemble de programmes informatiques pour le calcul des métriques d'évaluation.Deuxièmement, nous introduisons une nouvelle base de films de cinéma annotés, appelée Inria Action Movies, constitué de films d'action hollywoodiens, dont les plans sont annotés suivant des catégories sémantiques non-exclusives, dont la définition est suffisamment large pour couvrir l'ensemble du film. Un exemple de catégorie est "course-poursuite"; un autre exemple est "scène sentimentale". Nous proposons une approche pour localiser les sections de vidéos appartenant à chaque catégorie et apprendre les dépendances temporelles entre les occurrences de chaque catégorie.Troisièmement, nous décrivons les différentes versions du système développé pour la compétition de détection d'événement vidéo TRECVID Multimédia Event Detection, entre 2011 et 2014, en soulignant les composantes du système dont l'auteur du manuscrit était responsableAutomatic interpretation and understanding of videos still remains at the frontier of computer vision. The core challenge is to lift the expressive power of the current visual features (as well as features from other modalities, such as audio or text) to be able to automatically recognize typical video sections, with low temporal saliency yet high semantic expression. Examples of such long events include video sections where someone is fishing (TRECVID Multimedia Event Detection), or where the hero argues with a villain in a Hollywood action movie (Inria Action Movies). In this manuscript, we present several contributions towards this goal, focusing on three video analysis tasks: summarization, classification, localisation.First, we propose an automatic video summarization method, yielding a short and highly informative video summary of potentially long videos, tailored for specified categories of videos. We also introduce a new dataset for evaluation of video summarization methods, called MED-Summaries, which contains complete importance-scorings annotations of the videos, along with a complete set of evaluation tools.Second, we introduce a new dataset, called Inria Action Movies, consisting of long movies, and annotated with non-exclusive semantic categories (called beat-categories), whose definition is broad enough to cover most of the movie footage. Categories such as "pursuit" or "romance" in action movies are examples of beat-categories. We propose an approach for localizing beat-events based on classifying shots into beat-categories and learning the temporal constraints between shots.Third, we overview the Inria event classification system developed within the TRECVID Multimedia Event Detection competition and highlight the contributions made during the work on this thesis from 2011 to 2014
45
Yang, Guoli. "Learning in adaptive networks : analytical and computational approaches." Thesis, University of Edinburgh, 2016. http://hdl.handle.net/1842/20956.
Full textAbstract:
The dynamics on networks and the dynamics of networks are usually entangled with each other in many highly connected systems, where the former means the evolution of state and the latter means the adaptation of structure. In this thesis, we will study the coupled dynamics through analytical and computational approaches, where the adaptive networks are driven by learning of various complexities. Firstly, we investigate information diffusion on networks through an adaptive voter model, where two opinions are competing for the dominance. Two types of dynamics facilitate the agreement between neighbours: one is pairwise imitation and the other is link rewiring. As the rewiring strength increases, the network of voters will transform from consensus to fragmentation. By exploring various strategies for structure adaptation and state evolution, our results suggest that network configuration is highly influenced by range-based rewiring and biased imitation. In particular, some approximation techniques are proposed to capture the dynamics analytically through moment-closure differential equations. Secondly, we study an evolutionary model under the framework of natural selection. In a structured community made up of cooperators and cheaters (or defectors), a new-born player will adopt a strategy and reorganise its neighbourhood based on social inheritance. Starting from a cooperative population, an invading cheater may spread in the population occasionally leading to the collapse of cooperation. Such a collapse unfolds rapidly with the change of external conditions, bearing the traits of a critical transition. In order to detect the risk of invasions, some indicators based on population composition and network structure are proposed to signal the fragility of communities. Through the analyses of consistency and accuracy, our results suggest possible avenues for detecting the loss of cooperation in evolving networks. Lastly, we incorporate distributed learning into adaptive agents coordination, which emerges as a consequence of rational individual behaviours. A generic framework of work-learn-adapt (WLA) is proposed to foster the success of agents organisation. To gain higher organisation performance, the division of labour is achieved by a series of events of state evolution and structure adaptation. Importantly, agents are able to adjust their states and structures through quantitative information obtained from distributed learning. The adaptive networks driven by explicit learning pave the way for a better understanding of intelligent organisations in real world.
46
SEGUY, Vivien Pierre François. "Measure Transport Approaches for Data Visualization and Learning." Kyoto University, 2018. http://hdl.handle.net/2433/233857.
Full text
47
Лещенко, Ольга Іллівна, Ольга Ильинична Лещенко, and Olha Illivna Leshchenko. "Effective Training Approaches to Learning/Teaching Business English." Thesis, Sumy State University, 2017. http://essuir.sumdu.edu.ua/handle/123456789/67270.
Full textAbstract:
English language teaching has gone through a radical shift of
emphasis in the last forty years, beginning with the communicative
revolution in the mid-1970-s. The English language teachers are no longer
all alone in their role as trainers as they are able to bring parts of the
outside world to the classroom. Beyond the trainer’s “chalk and talk” and
the trainees’ language and communication practice there is a range of
materials in different media, which help the trainer to turn the Business
English classroom into a varied learning/teaching environment. While
printed text remains the most common training medium, both the ELT
world and the non-ELT world give us access to a wide range of video and
audio material, and increasingly multimedia.
48
Rarey, Margaret Shaker. "Observing and identifying young children's approaches to learning /." View abstract, 1999. http://library.ctstateu.edu/ccsu%5Ftheses/1574.html.
Full textAbstract:
Thesis (M.S.)--Central Connecticut State University, 1999.Thesis advisor: Claudia Shuster. " ... in partial fulfillment of the requirements for the degree of Master of Science [in Elementary and Early Childhood Education]." Includes bibliographical references (leaves 96-97).
49
Rajapakshage, N. (Nuwanthika). "Potential deep learning approaches for the physical layer." Master's thesis, University of Oulu, 2019. http://jultika.oulu.fi/Record/nbnfioulu-201908142760.
Full textAbstract:
Abstract. Deep learning based end-to-end learning of a communications system tries to optimize both transmitter and receiver blocks in a single process in an end-to-end manner, eliminating the need for artificial block structure of the conventional communications systems. Recently proposed concept of autoencoder based end-to-end communications is investigated in this thesis to validate its potential as an alternative to conventional block structured communications systems. A single user scenario in the additive white Gaussian noise (AWGN) channel is considered in this thesis. Autoencoder based systems are implemented equivalent to conventional communications systems and bit error rate (BER) performances of both systems are compared in different system settings.
Simulations show that the autoencoder outperforms equivalent uncoded binary phase shift keying (BPSK) system with a 2 dB margin to BPSK for a BER of 10⁻⁵, and has comparable performance to uncoded quadrature phase shift keying (QPSK) system. Autoencoder implementations equivalent to coded BPSK have shown comparable BER performance to hard decision convolutional coding (CC) with less than 1 dB gap over the 0–10 dB Eb/N0 range. Autoencoder is observed to have close performance to the conventional systems for higher code rates. Newly proposed autoencoder model as an alternative to coded systems with higher order modulations has shown that autoencoder is capable of learning better transmission mechanisms compared to the conventional systems adhering to the system parameters and resource constraints provided. Autoencoder equivalent of half-rate 16-quadrature amplitude modulation (16-QAM) system achieves a better performance with respect to hard decision CC over the 0–10 dB Eb/N0 range, and a comparable performance to soft decision CC with a better BER in 0–4 dB Eb/N0.
Comparable BER performance, lower processing complexity and low latency processing due to inherent parallel processing architecture, flexible structure and higher learning capacity are identified as advantages of the autoencoder based systems which show their potential and feasibility as an alternative to conventional communications systems.
50
Ouyang, Li. "Motivation, cultural values, learning processes, and learning in Chinese students." Thesis, Kingston, Ont. : [s.n.], 2008. http://hdl.handle.net/1974/1340.
Full text