Global ETD Search

11	APPLIED DEEP REINFORCEMENT LEARNING IN SMART ENERGY SYSTEMS MANAGEMENT Moein Sabounchi (17565402) 07 December 2023 (has links) <p dir="ltr">The horizon for inclusion of data-driven algorithms in cyber-physical systems is rapidly expanding due to evermore availability of high-performance computing tools and the inception of novel mathematical models in the fields of deep learning and reinforcement learning. In this regard, energy systems are a suitable candidate for data-driven algorithms utilization due to rapid expansion of smart measuring tools and infrastructure. Accordingly, I decided to explore the capabilities of deep reinforcement learning in control, security, and restoration of smart energy systems to tackle well-known problems such as ensuring stability, adversarial attack avoidance, and the black start restoration. To achieve this goal, I employed various reinforcement learning techniques in different capacities to develop transfer learning modules based on a rule-based approach for online control of the power system, utilized reinforcement learning for procedural noise generation in adversarial attacks against contingency detection in a power system and exploited multiple reinforcement learning algorithms to fully restore an energy system in an optimal manner. Per the results of these endeavors, I managed to develop a rule-based transfer learning logic to control the power system under various disturbance types and intensities. Furthermore, I developed an optimal adversarial attack module using a reinforcement-learning-based procedural noise generation to avoid detection by conventional deep-learning-based detection. Finally for the system restoration, the proposed intelligent restoration module managed to provide sustainable results for the black start restoration in energy system.</p> Planning and decision making Deep learning Reinforcement learning Artificial Intelligence Reinforcement Learning Smart energy systems Restoration AI Decision-Making
12	<b>WEARABLE BIG DATA HARNESSING WITH DEEP LEARNING, EDGE COMPUTING AND EFFICIENCY OPTIMIZATION</b> Jiadao Zou (16920153) 03 January 2024 (has links) <p dir="ltr">In this dissertation, efforts and innovations are made to advance subtle pattern mining, edge computing, and system efficiency optimization for biomedical applications, thereby advancing precision medicine big data.</p><p dir="ltr">Brain visual dynamics encode rich functional and biological patterns of the neural system, promising for applications like intention decoding, cognitive load quantization and neural disorder measurement. We here focus on the understanding of the brain visual dynamics for the Amyotrophic lateral sclerosis (ALS) population. We leverage a deep learning framework for automatic feature learning and classification, which can translate the eye Electrooculography (EOG) signal to meaningful words. We then build an edge computing platform on the smart phone, for learning, visualization, and decoded word demonstration, all in real-time. In a further study, we have leveraged deep transfer learning to boost EOG decoding effectiveness. More specifically, the model trained on basic eye movements is leveraged and treated as an additional feature extractor when classifying the signal to the meaningful word, resulting in higher accuracy.</p><p dir="ltr">Efforts are further made to decoding functional Near-Infrared Spectroscopy (fNIRS) signal, which encodes rich brain dynamics like the cognitive load. We have proposed a novel Multi-view Multi-channel Graph Neural Network (mmGNN). More specifically, we propose to mine the multi-channel fNIRS dynamics with a multi-stage GNN that can effectively extract the channel- specific patterns, propagate patterns among channels, and fuse patterns for high-level abstraction. Further, we boost the learning capability with multi-view learning to mine pertinent patterns in temporal, spectral, time-frequency, and statistical domains.</p><p dir="ltr">Massive-device systems, like wearable massive-sensor computers and Internet of Things (IoTs), are promising in the era of big data. The crucial challenge is about how to maximize the efficiency under coupling constraints like energy budget, computing, and communication. We propose a deep reinforcement learning framework, with a pattern booster and a learning adaptor. This framework has demonstrated optimally maximizes the energy utilization and computing efficiency on the local massive devices under a one-center fifteen-device circumstance.</p><p dir="ltr">Our research and findings are expected to greatly advance the intelligent, real-time, and efficient big data harnessing, leveraging deep learning, edge computing, and efficiency optimization.</p> Biomedical imaging Applications in health Planning and decision making Computational imaging Pattern recognition Human-computer interaction Machine Learning, Deep Learning smart healthcare wearables Health (medical Transfer Learning, Edge computation
13	Learning From Data Across Domains: Enhancing Human and Machine Understanding of Data From the Wild Sean Michael Kulinski (17593182) 13 December 2023 (has links) <p dir="ltr">Data is collected everywhere in our world; however, it often is noisy and incomplete. Different sources of data may have different characteristics, quality levels, or come from dynamic and diverse environments. This poses challenges for both humans who want to gain insights from data and machines which are learning patterns from data. How can we leverage the diversity of data across domains to enhance our understanding and decision-making? In this thesis, we address this question by proposing novel methods and applications that use multiple domains as more holistic sources of information for both human and machine learning tasks. For example, to help human operators understand environmental dynamics, we show the detection and localization of distribution shifts to problematic features, as well as how interpretable distributional mappings can be used to explain the differences between shifted distributions. For robustifying machine learning, we propose a causal-inspired method to find latent factors that are robust to environmental changes and can be used for counterfactual generation or domain-independent training; we propose a domain generalization framework that allows for fast and scalable models that are robust to distribution shift; and we introduce a new dataset based on human matches in StarCraft II that exhibits complex and shifting multi-agent behaviors. We showcase our methods across various domains such as healthcare, natural language processing (NLP), computer vision (CV), etc. to demonstrate that learning from data across domains can lead to more faithful representations of data and its generating environments for both humans and machines.</p> Knowledge representation and reasoning Natural language processing Planning and decision making Data engineering and data science Data mining and knowledge discovery Stream and sensor data Human-computer interaction Mixed initiative and human-in-the-loop Machine Learning Distribution Shifts Domain Generalization Artificial Intelligence
14	The Participation of Marginalized Populations in Health Services Planning and Decision Making Montesanti, Rose Stephanie 04 1900 (has links) <p>Community participation has been identified as a key facilitator of community health among marginalized populations in international health statements. However, knowledge gaps in the community participation literature regarding marginalized populations has been attributed to the lack of consistent definitions of community participation, ambiguity about the features of community participation initiatives (e.g., methods and strategies) that are appropriate for marginalized populations, and limitations of existing community participation frameworks in specifying the ways and means in which different marginalized populations might effectively participate, as well as in recognizing that community participation is highly contextual and situational. All of these factors have made it difficult to draw broader conclusions about the impact of participation methods and strategies for marginalized populations from evaluations of participation initiatives.</p> <p>The overall purpose of this thesis is to better understand how to involve marginalized populations in the planning and decision-making for local health services. First, a critical interpretive synthesis (CIS) was conducted to better understand the role of community development principles used in community participation initiatives with marginalized populations and the factors contributing to the influence of the principles in enabling the participation of these populations. Second, an in-depth comparative case study of four community participation initiatives in Ontario Community Health Centres (CHCs)—which are primary health care organizations serving 74 high-risk communities throughout the Province of Ontario—was conducted to identify the core features of participation initiatives with marginalized populations, and reflect on the particular challenges of engaging marginalized populations. Third, four focus groups were held at four Ontario CHCs to examine the role of frameworks as mechanisms for translating knowledge about community participation practice with marginalized populations. Overall, this thesis broadens our understanding of community participation with marginalized populations in the context of local health services planning and decision making. Specifically, this thesis contributes a theoretical basis for future research and provides practical knowledge for planning and evaluating community participation initiatives with marginalized populations.</p> / Doctor of Philosophy (PhD) community participation marginalized populations Community-based Research Community Engagement Community Health Health Services Research Inequality and Stratification Knowledge Translation Social and Cultural Anthropology Work, Economy and Organizations Community-based Research
15	Towards Novelty-Resilient AI: Learning in the Open World Trevor A Bonjour (18423153) 22 April 2024 (has links) <p dir="ltr">Current artificial intelligence (AI) systems are proficient at tasks in a closed-world setting where the rules are often rigid. However, in real-world applications, the environment is usually open and dynamic. In this work, we investigate the effects of such dynamic environments on AI systems and develop ways to mitigate those effects. Central to our exploration is the concept of \textit{novelties}. Novelties encompass structural changes, unanticipated events, and environmental shifts that can confound traditional AI systems. We categorize novelties based on their representation, anticipation, and impact on agents, laying the groundwork for systematic detection and adaptation strategies. We explore novelties in the context of stochastic games. Decision-making in stochastic games exercises many aspects of the same reasoning capabilities needed by AI agents acting in the real world. A multi-agent stochastic game allows for infinitely many ways to introduce novelty. We propose an extension of the deep reinforcement learning (DRL) paradigm to develop agents that can detect and adapt to novelties in these environments. To address the sample efficiency challenge in DRL, we introduce a hybrid approach that combines fixed-policy methods with traditional DRL techniques, offering enhanced performance in complex decision-making tasks. We present a novel method for detecting anticipated novelties in multi-agent games, leveraging information theory to discern patterns indicative of collusion among players. Finally, we introduce DABLER, a pioneering deep reinforcement learning architecture that dynamically adapts to changing environmental conditions through broad learning approaches and environment recognition. Our findings underscore the importance of developing AI systems equipped to navigate the uncertainties of the open world, offering promising pathways for advancing AI research and application in real-world settings.</p> Autonomous agents and multiagent systems Knowledge representation and reasoning Modelling and simulation Planning and decision making Context learning Deep learning Neural networks Reinforcement learning Novelty Context MDP CMDP Reinforcement Learning Deep Reinforcement Learning Collusion Novelty Detection Detection Adaptation
16	當大學遇到市場: 中國兩所大學新增本科專業的研究. / When universities meet market: emergence of new udergraduate programs in two Chinese universities / Emergence of new udergraduate programs in two Chinese universities / 中國兩所大學新增本科專業的研究 / CUHK electronic theses & dissertations collection / Dang da xue yu dao shi chang: Zhongguo liang suo da xue xin zeng ben ke zhuan ye de yan jiu. / Zhongguo liang suo da xue xin zeng ben ke zhuan ye de yan jiu January 2006 (has links) 陳霜叶. / 論文(哲學博士)--香港中文大學, 2006. / 參考文獻(p. 258-270). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in Chinese and English. / Lun wen (zhe xue bo shi)--Xianggang Zhong wen da xue, 2006. / Can kao wen xian (p. 258-270). / Chen Shuangye. Curriculum planning--Decision-making Education, Higher--Curricula Education, Higher--Economic aspects Education, Higher--Social aspects Higher education and state Universities and colleges--Curricula
17	DISTRIBUTED MACHINE LEARNING OVER LARGE-SCALE NETWORKS Frank Lin (16553082) 18 July 2023 (has links) <p>The swift emergence and wide-ranging utilization of machine learning (ML) across various industries, including healthcare, transportation, and robotics, have underscored the escalating need for efficient, scalable, and privacy-preserving solutions. Recognizing this, we present an integrated examination of three novel frameworks, each addressing different aspects of distributed learning and privacy issues: Two Timescale Hybrid Federated Learning (TT-HF), Delay-Aware Federated Learning (DFL), and Differential Privacy Hierarchical Federated Learning (DP-HFL). TT-HF introduces a semi-decentralized architecture that combines device-to-server and device-to-device (D2D) communications. Devices execute multiple stochastic gradient descent iterations on their datasets and sporadically synchronize model parameters via D2D communications. A unique adaptive control algorithm optimizes step size, D2D communication rounds, and global aggregation period to minimize network resource utilization and achieve a sublinear convergence rate. TT-HF outperforms conventional FL approaches in terms of model accuracy, energy consumption, and resilience against outages. DFL focuses on enhancing distributed ML training efficiency by accounting for communication delays between edge and cloud. It also uses multiple stochastic gradient descent iterations and periodically consolidates model parameters via edge servers. The adaptive control algorithm for DFL mitigates energy consumption and edge-to-cloud latency, resulting in faster global model convergence, reduced resource consumption, and robustness against delays. Lastly, DP-HFL is introduced to combat privacy vulnerabilities in FL. Merging the benefits of FL and Hierarchical Differential Privacy (HDP), DP-HFL significantly reduces the need for differential privacy noise while maintaining model performance, exhibiting an optimal privacy-performance trade-off. Theoretical analysis under both convex and nonconvex loss functions confirms DP-HFL’s effectiveness regarding convergence speed, privacy performance trade-off, and potential performance enhancement with appropriate network configuration. In sum, the study thoroughly explores TT-HF, DFL, and DP-HFL, and their unique solutions to distributed learning challenges such as efficiency, latency, and privacy concerns. These advanced FL frameworks have considerable potential to further enable effective, efficient, and secure distributed learning.</p> Network engineering Signal processing Knowledge representation and reasoning Modelling and simulation Planning and decision making Numerical analysis Optimisation Federated Learning frameworks differential privacy composition network optimization algorithm distributed machine learning (ML) Convergence analysis Edge Intelligence edge cloud computing
18	Trustworthy AI: Ensuring Explainability and Acceptance Davinder Kaur (17508870) 03 January 2024 (has links) <p dir="ltr">In the dynamic realm of Artificial Intelligence (AI), this study explores the multifaceted landscape of Trustworthy AI with a dedicated focus on achieving both explainability and acceptance. The research addresses the evolving dynamics of AI, emphasizing the essential role of human involvement in shaping its trajectory.</p><p dir="ltr">A primary contribution of this work is the introduction of a novel "Trustworthy Explainability Acceptance Metric", tailored for the evaluation of AI-based systems by field experts. Grounded in a versatile distance acceptance approach, this metric provides a reliable measure of acceptance value. Practical applications of this metric are illustrated, particularly in a critical domain like medical diagnostics. Another significant contribution is the proposal of a trust-based security framework for 5G social networks. This framework enhances security and reliability by incorporating community insights and leveraging trust mechanisms, presenting a valuable advancement in social network security.</p><p dir="ltr">The study also introduces an artificial conscience-control module model, innovating with the concept of "Artificial Feeling." This model is designed to enhance AI system adaptability based on user preferences, ensuring controllability, safety, reliability, and trustworthiness in AI decision-making. This innovation contributes to fostering increased societal acceptance of AI technologies. Additionally, the research conducts a comprehensive survey of foundational requirements for establishing trustworthiness in AI. Emphasizing fairness, accountability, privacy, acceptance, and verification/validation, this survey lays the groundwork for understanding and addressing ethical considerations in AI applications. The study concludes with exploring quantum alternatives, offering fresh perspectives on algorithmic approaches in trustworthy AI systems. This exploration broadens the horizons of AI research, pushing the boundaries of traditional algorithms.</p><p dir="ltr">In summary, this work significantly contributes to the discourse on Trustworthy AI, ensuring both explainability and acceptance in the intricate interplay between humans and AI systems. Through its diverse contributions, the research offers valuable insights and practical frameworks for the responsible and ethical deployment of AI in various applications.</p> Autonomous agents and multiagent systems Modelling and simulation Planning and decision making Collaborative and social computing Human-computer interaction Mixed initiative and human-in-the-loop Trustworthy AI Responsible AI explainability AI (XAI) AI Explainability Acceptance Metrics AI Acceptance Causal Inference Model AI Conscience AI Control
19	Machine-aided Bridge Vulnerability and Condition Management Xin Zhang (18364206) 15 April 2024 (has links) <p dir="ltr">Bridge safety has been a longstanding priority for civil engineers. Engineers and researchers devote significant effort toward establishing the ability to detect and monitor damage and to manage the performance of our bridges over their lifecycle. However, current practices, heavily reliant on manual involvement, still present challenges in efficiency and effectiveness. For instance, rapid bridge vulnerability assessment methods are being developed, but these methods normally require information about the bridge, e.g., substructure type, which is not readily available in most current bridge databases (e.g., National Bridge Inventory). Manually collecting the necessary information for each bridge is time-consuming and would influence the use of those rapid bridge vulnerability assessment methods. Similarly, routine bridge inspection, mandated every two years in the United States, requires organizing inspection photos, evaluating bridge condition, etc., which are also time-consuming tasks. A two-year inspection cycle may be overly cautious, especially in the early stages of its life when there is typically little degradation. Furthermore, when conducting an inspection there is no reference for when to use advanced inspection techniques, so visual inspection, the simplest method, is adopted for most bridges. Given these challenges in bridge asset management, the integration of machine learning to use the data from historical records of bridge performance can aid in and serve to expedite the above tasks.</p><p dir="ltr">The objective of this research is to develop machine learning-based methods that can assist humans in completing certain tasks associated with bridge asset management. Towards this objective, the following research tasks are carried out. In Task 1, a CNN-based bridge substructure identifier is developed to automatically recognize the bridge substructure type from an inspection image. A method is developed to set a rational budget for this work based on risk tolerance. In Task 2, the automated bridge inspection image organization tool (ABIRT) is developed to automate the process of organizing inspection images and generating an inspection report. In Task 3, a technique for machine-assisted bridge damage analysis using visual data is developed and validated. Additionally, a decision-making method is established to assist bridge inspectors in adopting this technique, with a focus on managing costs and minimizing risks. And in Task 4, reinforcement learning-based approach is developed to manage the bridge inspection process. Through this research, several key machine learning techniques are explored to assist bridge managers with the more tedious steps involved in the seismic vulnerability analysis and condition management tasks, allowing the engineer to dedicate more time to making decisions. This highly cross-disciplinary research is scalable and expandable to many other applications and will serve to improve the future safety and reliability of our infrastructure.</p> Earthquake engineering Structural engineering Transport engineering Planning and decision making Computer vision Image processing Deep learning Neural networks Reinforcement learning damage evaluation and classification Deep Learning Reinforcement Learning Artificial Intelligence Machine-aided Infrastructure Evaluation Decision-making and Plan Computer Vision in Civil Engineering seismic vulnerability analysis Human-machine collaboration physics-based analysis
20	EXPLORING GRAPH NEURAL NETWORKS FOR CLUSTERING AND CLASSIFICATION Fattah Muhammad Tahabi (14160375) 03 February 2023 (has links) <p><strong>Graph Neural Networks</strong> (GNNs) have become excessively popular and prominent deep learning techniques to analyze structural graph data for their ability to solve complex real-world problems. Because graphs provide an efficient approach to contriving abstract hypothetical concepts, modern research overcomes the limitations of classical graph theory, requiring prior knowledge of the graph structure before employing traditional algorithms. GNNs, an impressive framework for representation learning of graphs, have already produced many state-of-the-art techniques to solve node classification, link prediction, and graph classification tasks. GNNs can learn meaningful representations of graphs incorporating topological structure, node attributes, and neighborhood aggregation to solve supervised, semi-supervised, and unsupervised graph-based problems. In this study, the usefulness of GNNs has been analyzed primarily from two aspects - <strong>clustering and classification</strong>. We focus on these two techniques, as they are the most popular strategies in data mining to discern collected data and employ predictive analysis.</p> Biomechanical engineering Neural engineering Health promotion Preventative health care Applications in health Spatial data and applications Evolutionary computation Natural language processing Planning and decision making Data engineering and data science Data mining and knowledge discovery Graph, social and multimedia data Information retrieval and web search Knowledge and information management Context learning Deep learning Neural networks Semi- and unsupervised learning Data structures and algorithms Graph neural network Node classification Graph clustering Temporal graphs dynamic graphs NODE2VEC Graph Attention Mechanism Hunting BiLSTM model EHR data colorectal Cancer Cancers Cancer symptoms symptom Symptom cluster studies Coauthorship networks network analysis Word2vec Hierarchical Clustering method Dunn index semantic analysis text mining Natural Language Processing Tool UMLS identifiers umls Clinical Data Management

Search results