Gotowe bibliografie tematyczne / Data-efficient Deep Learning

Spis treści

Artykuły w czasopismach
Rozprawy doktorskie
Książki
Części książek
Streszczenia konferencji

Gotowa bibliografia na temat „Data-efficient Deep Learning”

Autor: Grafiati

Data publikacji: 6 września 2023

Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych

Wybierz rodzaj źródła:

Zobacz listy aktualnych artykułów, książek, rozpraw, streszczeń i innych źródeł naukowych na temat „Data-efficient Deep Learning”.

Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.

Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.

Artykuły w czasopismach na temat "Data-efficient Deep Learning"

Chaudhary, Dr Sumit, Ms Neha Singh i Salaiya Pankaj. "Time-Efficient Algorithm for Data Annotation using Deep Learning". Indian Journal of Artificial Intelligence and Neural Networking 2, nr 5 (30.08.2022): 8–11. http://dx.doi.org/10.54105/ijainn.e1058.082522.

Pełny tekst źródła

Streszczenie:

Current generation emphasis on the Digital world which creates a lot of unbeneficial data. The paper is about data annotation using deep learning as there is a lot of data available online but which data is useful can be labeled using these techniques. The unstructured data is labeled by many techniques but implementation of deep learning for labeling the unstructured data results in saving the time with high efficiency. In this paper introduce a method for data annotation, for that we can use unlabeled data as input and it is classified using the K-Nearest Neighbor algorithm. K-Nearest Neighbor is the fastest and its accuracy is very high compared to other classification algorithms. After classified the unlabeled data we use it as input and annotate data using deep learning techniques. In deep learning we use an auto annotator for annotating data. After annotating data, check the accuracy of annotated data and time efficiency of data annotation. In case the accuracy is low then we can retrain the data and make it more accurate.

Style APA, Harvard, Vancouver, ISO itp.

Biswas, Surojit, Grigory Khimulya, Ethan C. Alley, Kevin M. Esvelt i George M. Church. "Low-N protein engineering with data-efficient deep learning". Nature Methods 18, nr 4 (kwiecień 2021): 389–96. http://dx.doi.org/10.1038/s41592-021-01100-y.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Edstrom, Jonathon, Yifu Gong, Dongliang Chen, Jinhui Wang i Na Gong. "Data-Driven Intelligent Efficient Synaptic Storage for Deep Learning". IEEE Transactions on Circuits and Systems II: Express Briefs 64, nr 12 (grudzień 2017): 1412–16. http://dx.doi.org/10.1109/tcsii.2017.2767900.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Feng, Wenhui, Chongzhao Han, Feng Lian i Xia Liu. "A Data-Efficient Training Method for Deep Reinforcement Learning". Electronics 11, nr 24 (16.12.2022): 4205. http://dx.doi.org/10.3390/electronics11244205.

Pełny tekst źródła

Streszczenie:

Data inefficiency is one of the major challenges for deploying deep reinforcement learning algorithms widely in industry control fields, especially in regard to long-horizon sparse reward tasks. Even in a simulation-based environment, it is often prohibitive to take weeks to train an algorithm. In this study, a data-efficient training method is proposed in which a DQN is used as a base algorithm, and an elaborate curriculum is designed for the agent in the simulation scenario to accelerate the training process. In the early stage of the training process, the distribution of the initial state is set close to the goal so the agent can obtain an informative reward easily. As the training continues, the initial state distribution is set farther from the goal for the agent to explore more state space. Thus, the agent can obtain a reasonable policy through fewer interactions with the environment. To bridge the sim-to-real gap, the parameters for the output layer of the neural network for the value function are fine-tuned. An experiment on UAV maneuver control is conducted in the proposed training framework to verify the method. We demonstrate that data efficiency is different for the same data in different training stages.

Style APA, Harvard, Vancouver, ISO itp.

Hu, Wenjin, Feng Liu i Jiebo Peng. "An Efficient Data Classification Decision Based on Multimodel Deep Learning". Computational Intelligence and Neuroscience 2022 (4.05.2022): 1–10. http://dx.doi.org/10.1155/2022/7636705.

Pełny tekst źródła

Streszczenie:

A single model is often used to classify text data, but the generalization effect of a single model on text data sets is poor. To improve the model classification accuracy, a method is proposed that is based on a deep neural network (DNN), recurrent neural network (RNN), and convolutional neural network (CNN) and integrates multiple models trained by a deep learning network architecture to obtain a strong text classifier. Additionally, to increase the flexibility and accuracy of the model, various optimizer algorithms are used to train data sets. Moreover, to reduce the interference in the classification results caused by stop words in the text data, data preprocessing and text feature vector representation are used before training the model to improve its classification accuracy. The final experimental results show that the proposed model fusion method can achieve not only improved classification accuracy but also good classification effects on a variety of data sets.

Style APA, Harvard, Vancouver, ISO itp.

Mairittha, Nattaya, Tittaya Mairittha i Sozo Inoue. "On-Device Deep Learning Inference for Efficient Activity Data Collection". Sensors 19, nr 15 (5.08.2019): 3434. http://dx.doi.org/10.3390/s19153434.

Pełny tekst źródła

Streszczenie:

Labeling activity data is a central part of the design and evaluation of human activity recognition systems. The performance of the systems greatly depends on the quantity and “quality” of annotations; therefore, it is inevitable to rely on users and to keep them motivated to provide activity labels. While mobile and embedded devices are increasingly using deep learning models to infer user context, we propose to exploit on-device deep learning inference using a long short-term memory (LSTM)-based method to alleviate the labeling effort and ground truth data collection in activity recognition systems using smartphone sensors. The novel idea behind this is that estimated activities are used as feedback for motivating users to collect accurate activity labels. To enable us to perform evaluations, we conduct the experiments with two conditional methods. We compare the proposed method showing estimated activities using on-device deep learning inference with the traditional method showing sentences without estimated activities through smartphone notifications. By evaluating with the dataset gathered, the results show our proposed method has improvements in both data quality (i.e., the performance of a classification model) and data quantity (i.e., the number of data collected) that reflect our method could improve activity data collection, which can enhance human activity recognition systems. We discuss the results, limitations, challenges, and implications for on-device deep learning inference that support activity data collection. Also, we publish the preliminary dataset collected to the research community for activity recognition.

Style APA, Harvard, Vancouver, ISO itp.

Duan, Yanjie, Yisheng Lv, Yu-Liang Liu i Fei-Yue Wang. "An efficient realization of deep learning for traffic data imputation". Transportation Research Part C: Emerging Technologies 72 (listopad 2016): 168–81. http://dx.doi.org/10.1016/j.trc.2016.09.015.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Sashank, Madipally Sai Krishna, Vijay Souri Maddila, Vikas Boddu i Y. Radhika. "Efficient deep learning based data augmentation techniques for enhanced learning on inadequate medical imaging data". ACTA IMEKO 11, nr 1 (31.03.2022): 6. http://dx.doi.org/10.21014/acta_imeko.v11i1.1226.

Pełny tekst źródła

Streszczenie:

<p class="Abstract">The world has come to a standstill with the Coronavirus taking over. In these dire times, there are fewer doctors and more patients and hence, treatment is becoming more and more difficult and expensive. In recent times, Computer Science, Machine Intelligence, measurement technology has made a lot of progress in the field of Medical Science hence aiding the automation of a lot of medical activities. One area of progress in this regard is the automation of the process of detection of respiratory diseases (such as COVID-19). There have been many Convolutional Neural Network (CNN) architectures and approaches that have been proposed for Chest X-Ray Classification. But a big problem still remains and that is the minimal availability of Medical X-Ray Images due to improper measurements. Due to this minimal availability of Chest X-Ray data, most CNN classifiers do not get trained to an optimal level and the required standards for automating the process are not met. In order to overcome this problem, we propose a new deep learning based approach for accurate measurements of physiological data.</p>

Style APA, Harvard, Vancouver, ISO itp.

Petrovic, Nenad, i Djordje Kocic. "Data-driven framework for energy-efficient smart cities". Serbian Journal of Electrical Engineering 17, nr 1 (2020): 41–63. http://dx.doi.org/10.2298/sjee2001041p.

Pełny tekst źródła

Streszczenie:

Energy management is one of the greatest challenges in smart cities. Moreover, the presence of autonomous vehicles makes this task even more complex. In this paper, we propose a data-driven smart grid framework which aims to make smart cities energy-efficient focusing on two aspects: energy trading and autonomous vehicle charging. The framework leverages deep learning, linear optimization, semantic technology, domain-specific modelling notation, simulation and elements of relay protection. The evaluation of deep learning module together with code generation time and energy distribution cost reduction performed within the simulation environment also presented in this paper are given. According to the results, the achieved energy distribution cost reduction varies and depends from case to case.

Style APA, Harvard, Vancouver, ISO itp.

Yue, Yang, Bingyi Kang, Zhongwen Xu, Gao Huang i Shuicheng Yan. "Value-Consistent Representation Learning for Data-Efficient Reinforcement Learning". Proceedings of the AAAI Conference on Artificial Intelligence 37, nr 9 (26.06.2023): 11069–77. http://dx.doi.org/10.1609/aaai.v37i9.26311.

Pełny tekst źródła

Streszczenie:

Deep reinforcement learning (RL) algorithms suffer severe performance degradation when the interaction data is scarce, which limits their real-world application. Recently, visual representation learning has been shown to be effective and promising for boosting sample efficiency in RL. These methods usually rely on contrastive learning and data augmentation to train a transition model, which is different from how the model is used in RL---performing value-based planning. Accordingly, the learned representation by these visual methods may be good for recognition but not optimal for estimating state value and solving the decision problem. To address this issue, we propose a novel method, called value-consistent representation learning (VCR), to learn representations that are directly related to decision-making. More specifically, VCR trains a model to predict the future state (also referred to as the "imagined state'') based on the current one and a sequence of actions. Instead of aligning this imagined state with a real state returned by the environment, VCR applies a Q value head on both of the states and obtains two distributions of action values. Then a distance is computed and minimized to force the imagined state to produce a similar action value prediction as that by the real state. We develop two implementations of the above idea for the discrete and continuous action spaces respectively. We conduct experiments on Atari 100k and DeepMind Control Suite benchmarks to validate their effectiveness for improving sample efficiency. It has been demonstrated that our methods achieve new state-of-the-art performance for search-free RL algorithms.

Style APA, Harvard, Vancouver, ISO itp.

Więcej źródeł

Rozprawy doktorskie na temat "Data-efficient Deep Learning"

Lundström, Dennis. "Data-efficient Transfer Learning with Pre-trained Networks". Thesis, Linköpings universitet, Datorseende, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-138612.

Pełny tekst źródła

Streszczenie:

Deep learning has dominated the computer vision field since 2012, but a common criticism of deep learning methods is their dependence on large amounts of data. To combat this criticism research into data-efficient deep learning is growing. The foremost success in data-efficient deep learning is transfer learning with networks pre-trained on the ImageNet dataset. Pre-trained networks have achieved state-of-the-art performance on many tasks. We consider the pre-trained network method for a new task where we have to collect the data. We hypothesize that the data efficiency of pre-trained networks can be improved through informed data collection. After exhaustive experiments on CaffeNet and VGG16, we conclude that the data efficiency indeed can be improved. Furthermore, we investigate an alternative approach to data-efficient learning, namely adding domain knowledge in the form of a spatial transformer to the pre-trained networks. We find that spatial transformers are difficult to train and seem to not improve data efficiency.

Style APA, Harvard, Vancouver, ISO itp.

Edstrom, Jonathon. "Embracing Visual Experience and Data Knowledge: Efficient Embedded Memory Design for Big Videos and Deep Learning". Diss., North Dakota State University, 2019. https://hdl.handle.net/10365/31558.

Pełny tekst źródła

Streszczenie:

Energy efficient memory designs are becoming increasingly important, especially for applications related to mobile video technology and machine learning. The growing popularity of smart phones, tablets and other mobile devices has created an exponential demand for video applications in today’s society. When mobile devices display video, the embedded video memory within the device consumes a large amount of the total system power. This issue has created the need to introduce power-quality tradeoff techniques for enabling good quality video output, while simultaneously enabling power consumption reduction. Similarly, power efficiency issues have arisen within the area of machine learning, especially with applications requiring large and fast computation, such as neural networks. Using the accumulated data knowledge from various machine learning applications, there is now the potential to create more intelligent memory with the capability for optimized trade-off between energy efficiency, area overhead, and classification accuracy on the learning systems. In this dissertation, a review of recently completed works involving video and machine learning memories will be covered. Based on the collected results from a variety of different methods, including: subjective trials, discovered data-mining patterns, software simulations, and hardware power and performance tests, the presented memories provide novel ways to significantly enhance power efficiency for future memory devices. An overview of related works, especially the relevant state-of-the-art research, will be referenced for comparison in order to produce memory design methodologies that exhibit optimal quality, low implementation overhead, and maximum power efficiency.
National Science Foundation
ND EPSCoR
Center for Computationally Assisted Science and Technology (CCAST)

Style APA, Harvard, Vancouver, ISO itp.

Sagen, Markus. "Large-Context Question Answering with Cross-Lingual Transfer". Thesis, Uppsala universitet, Institutionen för informationsteknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-440704.

Pełny tekst źródła

Streszczenie:

Models based around the transformer architecture have become one of the most prominent for solving a multitude of natural language processing (NLP)tasks since its introduction in 2017. However, much research related to the transformer model has focused primarily on achieving high performance and many problems remain unsolved. Two of the most prominent currently are the lack of high performing non-English pre-trained models, and the limited number of words most trained models can incorporate for their context. Solving these problems would make NLP models more suitable for real-world applications, improving information retrieval, reading comprehension, and more. All previous research has focused on incorporating long-context for English language models. This thesis investigates the cross-lingual transferability between languages when only training for long-context in English. Training long-context models in English only could make long-context in low-resource languages, such as Swedish, more accessible since it is hard to find such data in most languages and costly to train for each language. This could become an efficient method for creating long-context models in other languages without the need for such data in all languages or pre-training from scratch. We extend the models’ context using the training scheme of the Longformer architecture and fine-tune on a question-answering task in several languages. Our evaluation could not satisfactorily confirm nor deny if transferring long-term context is possible for low-resource languages. We believe that using datasets that require long-context reasoning, such as a multilingual TriviaQAdataset, could demonstrate our hypothesis’s validity.

Style APA, Harvard, Vancouver, ISO itp.

Nayak, Gaurav Kumar. "Data-efficient Deep Learning Algorithms for Computer Vision Applications". Thesis, 2022. https://etd.iisc.ac.in/handle/2005/6094.

Pełny tekst źródła

Streszczenie:

The performance of any deep learning model depends heavily on the quantity and quality of the available training data. The generalization of the trained deep models improves with the availability of a large number of training samples and hence these models are often referred to as ‘data-hungry’. However, large scale datasets may not always be available in practice due to proprietary/privacy reasons or because of the high cost of generation, annotation, transmission and storage of data. Hence, efficient utilization of available data is of utmost importance, and this gives rise to a class of ML problems, which is often referred to as “data-efficient deep learning”. In this thesis we study the various types of such problems for diverse applications in computer vision, where the aim is to design deep neural network-based solutions that do not rely on the availability of large quantities of training data to attain the desired performance goals. Under the aforementioned thematic area, this thesis focuses on three different scenarios, namely - (1) learning in the absence of training data, (2) learning with limited training data and (3) learning using selected subset of training data. Absence of training data: Pre-trained deep models hold their learnt knowledge in the form of model parameters that act as ‘memory’ for the trained models and help them generalize well on unseen data. In the first part of this thesis, we present solutions to a diverse set of ‘zero-shot’ tasks, where in absence of any training data (or even their statistics) the trained models are leveraged to synthesize data-representative samples. We dub them Data Impressions (DIs), which act as proxy to the training data. As the DIs are not tied to any specific application, we show their utility in solving several CV/ML tasks under the challenging data-free setup, such as unsupervised domain adaptation, continual learning as well as knowledge distillation (KD). We also study the adversarial robustness of lightweight models trained via knowledge distillation using DIs. Further, we demonstrate the efficacy of DIs in generating data-free Universal Adversarial Perturbations (UAPs) with better fooling rates. However, one limiting factor of this solution is the relatively high computation (i.e., several rounds of backpropagation) to synthesize each sample. In fact, the other natural alternatives such as GAN based solutions also suffer from similar computational overhead and complicated training procedures. This motivated us to explore the utility of target class-balanced ‘arbitrary’ data as transfer set, which achieves competitive distillation performance and can yield strong baselines for data-free KD. We have also proposed data-free solutions beyond classification by extending zero-shot distillation to the object detection task, where we compose the pseudo transfer set by synthesizing multi-object impressions from a pretrained faster RCNN model. Another concern with the deployment of given trained models is their vulnerability against adversarial attacks. The popular adversarial training strategies rely on availability of original training data or explicit regularization-based techniques. On the contrary, we propose test-time adversarial defense (detection and correction framework), which can provide robustness in absence of training data and their statistics. We observe significant improvements in adversarial accuracy with minimal drop in clean accuracy against state-of-the-art ‘Auto Attack’ without having to retrain the model. Further, we explore an even more challenging problem setup and make the first attempt to provide adversarial robustness to ‘black box’ models (i.e., model architecture, weights, training details are inaccessible) under a complete data-free set up. Our method minimizes adversarial contamination on perturbed samples via proposed ‘wavelet noise remover’ (WNR) that remove coefficients corresponding to high frequency components which are most likely to be corrupted by adversarial attack, and recovers the lost image content by training a ‘regenerator’ network. This results in a high boost in adversarial accuracy when WNR combined with the trained regenerator network is prepended to black box network. Limited training data: In the second part, we assume the availability of a few training samples, where access to trained models may or may not be provided. In the few-shot setup, existing works obtain robustness using sophisticated meta-learning techniques which rely on the generation of adversarial samples in every episode of training - thereby making it computationally expensive. We propose the first computationally cheaper non-meta learning approach for robust few-shot learning that does not require any adversarial sample. We perform pretraining using self-distillation to make the feature representation of low-frequency samples close to original samples of base classes. Similarly, we also improve the discriminability of low-frequency query set features that further boost the robustness. Our method obtains massive improvement in adversarial performance while being ≈5x faster compared to state-of-the-art adversarial meta-learning methods. However, empirical robustness methods do not guarantee robustness of the trained models against all the adversarial perturbations possible within a given threat model. Thus, we also propose a novel problem of certified robustness of pretrained models in limited data settings. Our method provides a novel sample-generation strategy that synthesize ‘boundary’ and ‘interpolated’ samples to augment the limited training data and uses them in training the denoiser (prepended to pretrained classifier) via aligning the feature representations at multiple granularities (both instance and distribution levels). We achieve significant improvements across diverse sample budgets and noise levels in the white-box and observe similar performance under challenging black-box setup. Selected subset of training data: In the third part, we enforce efficient utilization via intelligently doing selective sampling on existing training datasets to obtain representative samples for the target task such as distillation, incremental learning and person-reid. Adversarial attacks recently have shown robustness bias, where certain subgroups in a dataset (e.g. based on class, gender, etc.) are less robust than others. Existing works characterize a subgroup’s robustness bias by only checking individual sample’s proximity to the decision boundary. We propose a holistic approach for quantifying adversarial vulnerability of a sample by combining different perspectives and further develop a trustworthy system to alert the humans about the incoming samples that are highly likely to be misclassified. Moreover, we demonstrate the utility of the proposed metric for data (and time)-efficient knowledge distillation which achieves better performance compared to competing baselines. Other applications such as incremental learning and video based person-reid can also be framed as a subset selection problem where representative samples need to be selected. We leverage DPP (Determinantal Point Process) for choosing the relevant and diverse samples. In Incremental learning, we propose a new variant of k-DPP that uses the RBF kernel (termed as “RBF k-DPP”) for challenging task of animal pose estimation and further tackle class imbalance by using image warping as an augmentation technique to generate varied poses for a given image, leading to further gains in performance. In video based re-id, we propose SLGDPP method which exploits the sequential nature of the frames in video while avoiding noisy and redundant (correlated) frames, resulting in outperforming the baseline sampling methods.

Style APA, Harvard, Vancouver, ISO itp.

Afreen, Ahmad. "Data Efficient Domain Generalization". Thesis, 2022. https://etd.iisc.ac.in/handle/2005/6047.

Pełny tekst źródła

Streszczenie:

Deep neural networks has brought tremendous success in many areas of computer vision, such as image classification, retrieval, segmentation , etc. However, this success is mostly measured under two conditions namely (1) the underlying distribution of the test data is the same as the distribution of the data used for training the network and (2) The classes available for testing is the same as the one in training. These assumptions are very restrictive in nature and may not hold in real-life. Since new data categories are continuously being discovered, so it is important for the trained models to generalize to classes which has not been seen during training. Also, since the conditions under which the data is captured keeps on changing, so it is important for the trained model to generalize across unseen domains, which it has not encountered during training. Also, in general, the information about the class (whether it belongs to a seen or unseen class) or domain will not be known a-priori. Recently, researchers have started to address the challenging scenarios associated with a deep network, when the testing conditions in terms of classes and domains are relaxed. Towards that end, domain generalization (DG) for tasks like image classification, object detection, etc. have gained significant attention. In this work, we focus on the image classification task. In the first work, we address the scenario where the test data domain can be different and in the second work, we address the even more general scenario (ZSDG), where both the class and domain of the test data can be different from that of the training data. In DG, a deep model is trained to generalize well on an unknown target domain, leveraging data from multiple source domains during training for the task of image classification. In ZSDG, the aim is to train the model using multiple source domains and attributes of the classes such that it can generalize well to novel classes from out-of-distribution target data.

Style APA, Harvard, Vancouver, ISO itp.

Wong, Jun Hua. "Efficient Edge Intelligence in the Era of Big Data". Thesis, 2021. http://hdl.handle.net/1805/26385.

Pełny tekst źródła

Streszczenie:

Indiana University-Purdue University Indianapolis (IUPUI)
Smart wearables, known as emerging paradigms for vital big data capturing, have been attracting intensive attentions. However, one crucial problem is their power-hungriness, i.e., the continuous data streaming consumes energy dramatically and requires devices to be frequently charged. Targeting this obstacle, we propose to investigate the biodynamic patterns in the data and design a data-driven approach for intelligent data compression. We leverage Deep Learning (DL), more specifically, Convolutional Autoencoder (CAE), to learn a sparse representation of the vital big data. The minimized energy need, even taking into consideration the CAE-induced overhead, is tremendously lower than the original energy need. Further, compared with state-of-the-art wavelet compression-based method, our method can compress the data with a dramatically lower error for a similar energy budget. Our experiments and the validated approach are expected to boost the energy efficiency of wearables, and thus greatly advance ubiquitous big data applications in era of smart health. In recent years, there has also been a growing interest in edge intelligence for emerging instantaneous big data inference. However, the inference algorithms, especially deep learning, usually require heavy computation requirements, thereby greatly limiting their deployment on the edge. We take special interest in the smart health wearable big data mining and inference. Targeting the deep learning’s high computational complexity and large memory and energy requirements, new approaches are urged to make the deep learning algorithms ultra-efficient for wearable big data analysis. We propose to leverage knowledge distillation to achieve an ultra-efficient edge-deployable deep learning model. More specifically, through transferring the knowledge from a teacher model to the on-edge student model, the soft target distribution of the teacher model can be effectively learned by the student model. Besides, we propose to further introduce adversarial robustness to the student model, by stimulating the student model to correctly identify inputs that have adversarial perturbation. Experiments demonstrate that the knowledge distillation student model has comparable performance to the heavy teacher model but owns a substantially smaller model size. With adversarial learning, the student model has effectively preserved its robustness. In such a way, we have demonstrated the framework with knowledge distillation and adversarial learning can, not only advance ultra-efficient edge inference, but also preserve the robustness facing the perturbed input.

Style APA, Harvard, Vancouver, ISO itp.

Schwarzer, Max. "Data-efficient reinforcement learning with self-predictive representations". Thesis, 2020. http://hdl.handle.net/1866/25105.

Pełny tekst źródła

Streszczenie:

L'efficacité des données reste un défi majeur dans l'apprentissage par renforcement profond. Bien que les techniques modernes soient capables d'atteindre des performances élevées dans des tâches extrêmement complexes, y compris les jeux de stratégie comme le StarCraft, les échecs, le shogi et le go, ainsi que dans des domaines visuels exigeants comme les jeux Atari, cela nécessite généralement d'énormes quantités de données interactives, limitant ainsi l'application pratique de l'apprentissage par renforcement. Dans ce mémoire, nous proposons la SPR, une méthode inspirée des récentes avancées en apprentissage auto-supervisé de représentations, conçue pour améliorer l'efficacité des données des agents d'apprentissage par renforcement profond. Nous évaluons cette méthode sur l'environement d'apprentissage Atari, et nous montrons qu'elle améliore considérablement les performances des agents avec un surcroît de calcul modéré. Lorsqu'on lui accorde à peu près le même temps d'apprentissage qu'aux testeurs humains, un agent d'apprentissage par renforcement augmenté de SPR atteint des performances surhumaines dans 7 des 26 jeux, une augmentation de 350% par rapport à l'état de l'art précédent, tout en améliorant fortement les performances moyennes et médianes. Nous évaluons également cette méthode sur un ensemble de tâches de contrôle continu, montrant des améliorations substantielles par rapport aux méthodes précédentes. Le chapitre 1 présente les concepts nécessaires à la compréhension du travail présenté, y compris des aperçus de l'apprentissage par renforcement profond et de l'apprentissage auto-supervisé de représentations. Le chapitre 2 contient une description détaillée de nos contributions à l'exploitation de l'apprentissage de représentation auto-supervisé pour améliorer l'efficacité des données dans l'apprentissage par renforcement. Le chapitre 3 présente quelques conclusions tirées de ces travaux, y compris des propositions pour les travaux futurs.
Data efficiency remains a key challenge in deep reinforcement learning. Although modern techniques have been shown to be capable of attaining high performance in extremely complex tasks, including strategy games such as StarCraft, Chess, Shogi, and Go as well as in challenging visual domains such as Atari games, doing so generally requires enormous amounts of interactional data, limiting how broadly reinforcement learning can be applied. In this thesis, we propose SPR, a method drawing from recent advances in self-supervised representation learning designed to enhance the data efficiency of deep reinforcement learning agents. We evaluate this method on the Atari Learning Environment, and show that it dramatically improves performance with limited computational overhead. When given roughly the same amount of learning time as human testers, a reinforcement learning agent augmented with SPR achieves super-human performance on 7 out of 26 games, an increase of 350% over the previous state of the art, while also strongly improving mean and median performance. We also evaluate this method on a set of continuous control tasks, showing substantial improvements over previous methods. Chapter 1 introduces concepts necessary to understand the work presented, including overviews of Deep Reinforcement Learning and Self-Supervised Representation learning. Chapter 2 contains a detailed description of our contributions towards leveraging self-supervised representation learning to improve data-efficiency in reinforcement learning. Chapter 3 provides some conclusions drawn from this work, including a number of proposals for future work.

Style APA, Harvard, Vancouver, ISO itp.

(11013474), Jun Hua Wong. "Efficient Edge Intelligence In the Era of Big Data". Thesis, 2021.

Znajdź pełny tekst źródła

Streszczenie:

Smart wearables, known as emerging paradigms for vital big data capturing, have been attracting intensive attentions. However, one crucial problem is their power-hungriness, i.e., the continuous data streaming consumes energy dramatically and requires devices to be frequently charged. Targeting this obstacle, we propose to investigate the biodynamic patterns in the data and design a data-driven approach for intelligent data compression. We leverage Deep Learning (DL), more specifically, Convolutional Autoencoder (CAE), to learn a sparse representation of the vital big data. The minimized energy need, even taking into consideration the CAE-induced overhead, is tremendously lower than the original energy need. Further, compared with state-of-the-art wavelet compression-based method, our method can compress the data with a dramatically lower error for a similar energy budget. Our experiments and the validated approach are expected to boost the energy efficiency of wearables, and thus greatly advance ubiquitous big data applications in era of smart health.

In recent years, there has also been a growing interest in edge intelligence for emerging instantaneous big data inference. However, the inference algorithms, especially deep learning, usually require heavy computation requirements, thereby greatly limiting their deployment on the edge. We take special interest in the smart health wearable big data mining and inference.

Targeting the deep learning’s high computational complexity and large memory and energy requirements, new approaches are urged to make the deep learning algorithms ultra-efficient for wearable big data analysis. We propose to leverage knowledge distillation to achieve an ultra-efficient edge-deployable deep learning model. More specifically, through transferring the knowledge from a teacher model to the on-edge student model, the soft target distribution of the teacher model can be effectively learned by the student model. Besides, we propose to further introduce adversarial robustness to the student model, by stimulating the student model to correctly identify inputs that have adversarial perturbation. Experiments demonstrate that the knowledge distillation student model has comparable performance to the heavy teacher model but owns a substantially smaller model size. With adversarial learning, the student model has effectively preserved its robustness. In such a way, we have demonstrated the framework with knowledge distillation and adversarial learning can, not only advance ultra-efficient edge inference, but also preserve the robustness facing the perturbed input.

Style APA, Harvard, Vancouver, ISO itp.

Ehrler, Matthew. "VConstruct: a computationally efficient method for reconstructing satellite derived Chlorophyll-a data". Thesis, 2021. http://hdl.handle.net/1828/13346.

Pełny tekst źródła

Streszczenie:

The annual phytoplankton bloom is an important marine event. Its annual variability can be easily recognized by ocean-color satellite sensors through the increase in surface Chlorophyll-a concentration, a key indicator to quantitatively characterize all phytoplankton groups. However, a common problem is that the satellites used to gather the data are obstructed by clouds and other artifacts. This means that time series data from satellites can suffer from spatial data loss. There are a number of algorithms that are able to reconstruct the missing parts of these images to varying degrees of accuracy, with Data INterpolating Empirical Orthogonal Functions (DINEOF) being the most popular. However, DINEOF has a high computational cost, taking both significant time and memory to generate reconstructions. We propose a machine learning approach to reconstruction of Chlorophyll-a data using a Variational Autoencoder (VAE). Our method is 3-5x times faster (50-200x if the method has already been run once in the area). Our method uses less memory and increasing the size of the data being reconstructed causes computational cost to grow at a significantly better rate than DINEOF. We show that our method's accuracy is within a margin of error but slightly less accurate than DINEOF, as found by our own experiments and similar experiments from other studies. Lastly, we discuss other potential benefits of our method that could be investigated in future work, including generating data under certain conditions or anomaly detection.
Graduate

Style APA, Harvard, Vancouver, ISO itp.

Książki na temat "Data-efficient Deep Learning"

Jena, Om Prakash, Alok Ranjan Tripathy, Brojo Kishore Mishra i Ahmed A. Elngar, red. Augmented Intelligence: Deep Learning, Machine Learning, Cognitive Computing, Educational Data Mining. BENTHAM SCIENCE PUBLISHERS, 2022. http://dx.doi.org/10.2174/97898150404011220301.

Pełny tekst źródła

Streszczenie:

Blockchain, whether public or private, is capable enough to maintain the integrity of transactions by decentralizing the records for users. Many IoT companies are using blockchain technology to make the world a better-connected place. Businesses and researchers are exploring ways to make this technology increasingly efficient for IoT services. This volume presents the recent advances in these two technologies. Chapters explain the fundamentals of Blockchain and IoT, before explaining how these technologies, when merged together, provide a transparent, reliable, and secure model for data processing by intelligent devices in various domains. Readers will be able to understand how these technologies are making an impact on healthcare, supply chain management and electronic voting, to give a few examples. The 10 peer-reviewed book chapters have been contributed by scholars, researchers, academicians, and engineering professionals, and provide a comprehensive yet easily digestible update on Blockchain on IoT technology.

Style APA, Harvard, Vancouver, ISO itp.

Delgado Martín, Jordi, Andrea Muñoz-Ibáñez i Ismael Himar Falcón-Suárez. 6th International Workshop on Rock Physics: A Coruña, Spain 13 -17 June 2022: Book of Abstracts. Wyd. 2022. Servizo de Publicacións da UDC, 2022. http://dx.doi.org/10.17979/spudc.000005.

Pełny tekst źródła

Streszczenie:

[Abstract] The 6th International Workshop on Rock Physics (6IWRP) was held A Coruña, Spain, between 13th and 17th of June, 2022. This meeting follows the track of the five successful encounters held in Golden (USA, 2011), Southampton (UK, 2013), Perth (Australia, 2015), Trondheim (Norway, 2017) and Hong Kong (China, 2019). The aim of the workshop was to bring together experiences allowing to illustrate, discuss and exchange recent advances in the wide realm of rock physics, including theoretical developments, in situ and laboratory scale experiments as well as digital analysis. While rock physics is at the core of the oil & gas industry applications, it is also essential to enable the energy transition challenge (e.g. CO2 and H2 storage, geothermal), ensure a safe and adequate use of natural resources and develop efficient waste management strategies. The topics of 6IWRP covered a broad spectrum of rock physics-related research activities, including: • Experimental rock physics. New techniques, approaches and applications; Characterization of the static and dynamic properties of rocks and fluids; Multiphysics measurements (NMR, electrical resistivity…); Deep/crustal scale rock physics. • Modelling and multiscale applications: from the lab to the field. Numerical analysis and model development; Data science applications; Upscaling; Microseismicity and earthquakes; Subsurface stresses and tectonic deformations. • Coupled phenomena and rock properties: exploring interactions. Anisotropy; Flow and fractures; Temperature effects; Rock-fluid interaction; Fluid and pressure effects on geophysical signatures. • The energy transition challenge. Applications to energy storage (hydrogen storage in porous media), geothermal resources, energy production (gas hydrates), geological utilization and storage of CO2, nuclear waste disposal. • Rock physics templates: advances and applications. Quantitative assessment; Applications to reser voir characterization (role of seismic wave anisotropy and fracture networks). • Advanced rock physics tools. Machine learning; application of imaging (X-ray CT, X-ray μCT, FIB-SEM…) to obtain rock proper ties. This book compiles more than 50 abstracts, summarizing the works presented in the 6IWRP by rock physicists from all over the world, belonging to both academia and industry. This book means an updated overview of the rock physics research worldwide.

Style APA, Harvard, Vancouver, ISO itp.

Części książek na temat "Data-efficient Deep Learning"

Sarkar, Tirthajyoti. "Modular and Productive Deep Learning Code". W Productive and Efficient Data Science with Python, 113–56. Berkeley, CA: Apress, 2022. http://dx.doi.org/10.1007/978-1-4842-8121-5_5.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Vepakomma, Praneeth, i Ramesh Raskar. "Split Learning: A Resource Efficient Model and Data Parallel Approach for Distributed Deep Learning". W Federated Learning, 439–51. Cham: Springer International Publishing, 2022. http://dx.doi.org/10.1007/978-3-030-96896-0_19.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Maulana, Muhammad Rizki, i Wee Sun Lee. "Ensemble and Auxiliary Tasks for Data-Efficient Deep Reinforcement Learning". W Machine Learning and Knowledge Discovery in Databases. Research Track, 122–38. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-86486-6_8.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Ghantasala, G. S. Pradeep, L. R. Sudha, T. Veni Priya, P. Deepan i R. Raja Vignesh. "An Efficient Deep Learning Framework for Multimedia Big Data Analytics". W Multimedia Computing Systems and Virtual Reality, 99–127. Boca Raton: CRC Press, 2022. http://dx.doi.org/10.1201/9781003196686-5.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Sharma, Pranav, Marcus Rüb, Daniel Gaida, Heiko Lutz i Axel Sikora. "Deep Learning in Resource and Data Constrained Edge Computing Systems". W Machine Learning for Cyber Physical Systems, 43–51. Berlin, Heidelberg: Springer Berlin Heidelberg, 2020. http://dx.doi.org/10.1007/978-3-662-62746-4_5.

Pełny tekst źródła

Streszczenie:

AbstractTo demonstrate how deep learning can be applied to industrial applications with limited training data, deep learning methodologies are used in three different applications. In this paper, we perform unsupervised deep learning utilizing variational autoencoders and demonstrate that federated learning is a communication efficient concept for machine learning that protects data privacy. As an example, variational autoencoders are utilized to cluster and visualize data from a microelectromechanical systems foundry. Federated learning is used in a predictive maintenance scenario using the C-MAPSS dataset.

Style APA, Harvard, Vancouver, ISO itp.

Zheng, Yefeng, David Liu, Bogdan Georgescu, Hien Nguyen i Dorin Comaniciu. "Robust Landmark Detection in Volumetric Data with Efficient 3D Deep Learning". W Deep Learning and Convolutional Neural Networks for Medical Image Computing, 49–61. Cham: Springer International Publishing, 2017. http://dx.doi.org/10.1007/978-3-319-42999-1_4.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Symeonidis, C., P. Nousi, P. Tosidis, K. Tsampazis, N. Passalis, A. Tefas i N. Nikolaidis. "Efficient Realistic Data Generation Framework Leveraging Deep Learning-Based Human Digitization". W Proceedings of the International Neural Networks Society, 271–83. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-80568-5_23.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Rajawat, Anand Singh, Kanishk Barhanpurkar, S. B. Goyal, Pradeep Bedi, Rabindra Nath Shaw i Ankush Ghosh. "Efficient Deep Learning for Reforming Authentic Content Searching on Big Data". W Advanced Computing and Intelligent Technologies, 319–27. Singapore: Springer Singapore, 2021. http://dx.doi.org/10.1007/978-981-16-2164-2_26.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Zheng, Yefeng, David Liu, Bogdan Georgescu, Hien Nguyen i Dorin Comaniciu. "3D Deep Learning for Efficient and Robust Landmark Detection in Volumetric Data". W Lecture Notes in Computer Science, 565–72. Cham: Springer International Publishing, 2015. http://dx.doi.org/10.1007/978-3-319-24553-9_69.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Ghesu, Florin C., Bogdan Georgescu, Yefeng Zheng, Joachim Hornegger i Dorin Comaniciu. "Marginal Space Deep Learning: Efficient Architecture for Detection in Volumetric Image Data". W Lecture Notes in Computer Science, 710–18. Cham: Springer International Publishing, 2015. http://dx.doi.org/10.1007/978-3-319-24553-9_87.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Streszczenia konferencji na temat "Data-efficient Deep Learning"

Zhang, Xianchao, Wentao Yang, Xiaotong Zhang, Han Liu i Guanglu Wang. "Data-Efficient Deep Reinforcement Learning with Symmetric Consistency". W 2022 26th International Conference on Pattern Recognition (ICPR). IEEE, 2022. http://dx.doi.org/10.1109/icpr56361.2022.9956417.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Li, Hang, Ju Wang, Xi Chen, Xue Liu i Gregory Dudek. "Data-Efficient Communication Traffic Prediction With Deep Transfer Learning". W ICC 2022 - IEEE International Conference on Communications. IEEE, 2022. http://dx.doi.org/10.1109/icc45855.2022.9838413.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Wang, Xue. "An efficient federated learning optimization algorithm on non-IID data". W International Conference on Cloud Computing, Performance Computing, and Deep Learning (CCPCDL 2022), redaktor Sandeep Saxena. SPIE, 2022. http://dx.doi.org/10.1117/12.2640939.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Kok, Ibrahim, Burak H. Corak, Uraz Yavanoglu i Suat Ozdemir. "Deep Learning based Delay and Bandwidth Efficient Data Transmission in IoT". W 2019 IEEE International Conference on Big Data (Big Data). IEEE, 2019. http://dx.doi.org/10.1109/bigdata47090.2019.9005680.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Chen, Zhixin, Zhixin Jia, Mengxiang Lin i Shibo Jian. "Towards Generalization and Data Efficient Learning of Deep Robotic Grasping". W 2022 IEEE 17th Conference on Industrial Electronics and Applications (ICIEA). IEEE, 2022. http://dx.doi.org/10.1109/iciea54703.2022.10006045.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Xiong, Y., i J. Cheng. "Efficient Seismic Data Interpolation Using Deep Convolutional Networks and Transfer Learning". W 81st EAGE Conference and Exhibition 2019. European Association of Geoscientists & Engineers, 2019. http://dx.doi.org/10.3997/2214-4609.201900768.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Xu, Changming, Hengfeng Ding, Xuejian Zhang, Cong Wang i Hongji Yang. "A Data-Efficient Method of Deep Reinforcement Learning for Chinese Chess". W 2022 IEEE 22nd International Conference on Software Quality, Reliability, and Security Companion (QRS-C). IEEE, 2022. http://dx.doi.org/10.1109/qrs-c57518.2022.00109.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Wu, Di, Jikun Kang, Yi Tian Xu, Hang Li, Jimmy Li, Xi Chen, Dmitriy Rivkin i in. "Load Balancing for Communication Networks via Data-Efficient Deep Reinforcement Learning". W GLOBECOM 2021 - 2021 IEEE Global Communications Conference. IEEE, 2021. http://dx.doi.org/10.1109/globecom46510.2021.9685294.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Awad, Abdalaziz, Philipp Brendel i Andreas Erdmann. "Data efficient deep learning for imaging with novel EUV mask absorbers". W Optical and EUV Nanolithography XXXV, redaktorzy Anna Lio i Martin Burkhardt. SPIE, 2022. http://dx.doi.org/10.1117/12.2613954.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Melas-Kyriazi, Luke, George Han i Celine Liang. "Generation-Distillation for Efficient Natural Language Understanding in Low-Data Settings". W Proceedings of the 2nd Workshop on Deep Learning Approaches for Low-Resource NLP (DeepLo 2019). Stroudsburg, PA, USA: Association for Computational Linguistics, 2019. http://dx.doi.org/10.18653/v1/d19-6114.

Pełny tekst źródła

Style APA, Harvard, Vancouver, ISO itp.

Oferujemy zniżki na wszystkie plany premium dla autorów, których prace zostały uwzględnione w tematycznych zestawieniach literatury. Skontaktuj się z nami, aby uzyskać unikalny kod promocyjny!