Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 20 for “"Safe RL"”.

  1. Meta-optimization in safe reinforcement learning: Enhancing safety at training and deployment with fewer hyperparameters

    Reinforcement learning (RL) is a trial-and-error framework for enabling intelligent systems to learn the optimal behaviour based on the feedback from the environment. In recent years, successful application of RL in controlling various embodied systems have been observed. However, the real-world …

    uvic Repository record for Meta-optimization in safe reinforcement learning: Enhancing safety at training and deployment with fewer hyperparameters (opens in a new tab)

  2. JOINT LEARNING OF UNKNOWN SAFETY CONSTRAINTS AND CONTROL POLICIES IN REINFORCEMENT LEARNING

    Reinforcement learning (RL) has revolutionized decision-making across a wide range of domains over the past few decades. Yet, deploying RL policies in real-world scenarios presents the crucial challenge of ensuring safety. Traditional safe RL approaches have predominantly focused on incorporating …

    wvu Repository record for JOINT LEARNING OF UNKNOWN SAFETY CONSTRAINTS AND CONTROL POLICIES IN REINFORCEMENT LEARNING (opens in a new tab)

  3. Electric Bus Charging Scheduling based on Deep Reinforcement Learning

    … Hierarchical Deep Reinforcement Learning (HDRL) frameworks combined with Safe Reinforcement Learning (Safe RL) methods. The primary objective is to devise intelligent, scalable, and safety-aware charging scheduling algorithms tailored for practical, uncertain environments. Initially, we …

    guelph Repository record for Electric Bus Charging Scheduling based on Deep Reinforcement Learning (opens in a new tab)

  4. Towards AI Safety and Alignment in Interactive Environments

    … interaction (HCI) frameworks to enhance the safety and steerability of reinforcement learning (RL) models in complex game environments. Derived from the HCI perspective, the research emphasises capturing human-like behaviours, values, and interactions in AI systems, ensuring that machine …

    unsw Repository record for Towards AI Safety and Alignment in Interactive Environments (opens in a new tab) Embargoed

  5. Data-driven Supply Chain Monitoring and Optimization

    … under investigation is reinforcement learning (RL). Within the RL framework, an agent learns an optimal policy that maps the state to action during the process of interacting with the non-deterministic environment, such that a numerical reward is maximized. The current literature regarding …

    mcmaster Repository record for Data-driven Supply Chain Monitoring and Optimization (opens in a new tab)

  6. STATISTICAL METHODS FOR POLICY EVALUATION AND POLICY LEARNING

    … allocation and recursive Jensen MM for earlier stages via iterative backward sweeps to reduce MSE. Second, we propose a counterfactually safe RL framework, defining harm via potential outcomes and estimating it under causal-copula assumptions. A two-stage algorithm with fitted Q-iteration …

    nus Repository record for STATISTICAL METHODS FOR POLICY EVALUATION AND POLICY LEARNING (opens in a new tab) Download file for STATISTICAL METHODS FOR POLICY EVALUATION AND POLICY LEARNING (opens in a new tab)

  7. SAFE REINFORCEMENT LEARNING-BASED GREEN LIGHT OPTIMAL SPEED ADVISORY FOR MIXED-TRAFFIC PLATOONS

    This thesis develops a platoon-centric, safe RL-based Green Light Optimal Speed Advisory (GLOSA) system to optimize the CAV speed profile of a mixed-traffic platoon. First, we design a multi-agent RL algorithm to achieve a balance between the energy and travel efficiency of a mixed-traffic platoon, …

    nus Repository record for SAFE REINFORCEMENT LEARNING-BASED GREEN LIGHT OPTIMAL SPEED ADVISORY FOR MIXED-TRAFFIC PLATOONS (opens in a new tab) Download file for SAFE REINFORCEMENT LEARNING-BASED GREEN LIGHT OPTIMAL SPEED ADVISORY FOR MIXED-TRAFFIC PLATOONS (opens in a new tab)

  8. Provable Learning with Real-World Constraints

    … learning, and reinforcement learning (RL). These machine learning methods play an important role in many fields, such as data classification, data clustering, robot manipulation, etc. However, it is challenging to extend this success to real-world applications by directly applying …

    tamu Repository record for Provable Learning with Real-World Constraints (opens in a new tab)

  9. Reinforcement Learning Benchmarking for Sustainable Energy Systems: Perturbation Robustness, Safety Constraints, and Multi-Agent Coordination

    Reinforcement learning (RL) has emerged as a promising approach for sequential decision making in sustainable energy systems, yet the lack of systematic benchmarks hinders principled algorithm comparison and deployment. This thesis presents SustainRL-Bench, a comprehensive benchmarking study that …

    vt Repository record for Reinforcement Learning Benchmarking for Sustainable Energy Systems: Perturbation Robustness, Safety Constraints, and Multi-Agent Coordination (opens in a new tab)

  10. Bridging the gaps in trustworthy autonomy: Alignment, validation, and physical execution in learning-enabled cyber-physical systems

    … Among these, Reinforcement Learning (RL) has seen increasing adoption for sequential decision-making, offering a compelling alternative to supervised learning paradigms for handling dynamic control tasks. However, transferring these learning-based controllers to real-world hardware …

    iastate Repository record for Bridging the gaps in trustworthy autonomy: Alignment, validation, and physical execution in learning-enabled cyber-physical systems (opens in a new tab)

  11. Safe Reinforcement Learning Using Formally Verified Abstract Policies

    Reinforcement learning (RL) is an artificial intelligence technique for finding optimal solutions for sequential decision-making problems modelled as Markov decision processes (MDPs). Objectives are represented as numerical rewards in the model where positive values represent achievements and …

    whiterose Download PDF for Safe Reinforcement Learning Using Formally Verified Abstract Policies (opens in a new tab)

  12. Deep Reinforcement Learning for Autonomous Driving: Human-Informed, Ethical, and Transferable Agents

    … investigates how deep reinforcement learning (DRL) can be used to improve planning and control in autonomous driving (AD) systems. While DRL has shown promise in learning complex driving behaviors, practical deployment faces three persistent challenges: the sensitivity of agent behavior to …

    qucosa-diss

  13. Reinforcement Learning Based Motion Planner and Trajectory Tracker for Unmanned Aerial Systems

    … motivating the use of reinforcement learning (RL) for UAV control. This thesis explores RL-based UAV trajectory tracking through three key phases: benchmarking RL algorithms, developing a dual-agent RL framework, and leveraging these insights to design a model-free RL approach for aerial …

    waterloo Repository record for Reinforcement Learning Based Motion Planner and Trajectory Tracker for Unmanned Aerial Systems (opens in a new tab)

  14. On Achieving Acceptable Levels of Safety Risk in a Reinforcement Learning Environment

    … that model correctness does not ensure the safe operation of systems that perform safety-critical functions. However, to ensure that potential hazardous events cannot occur, the model must be verified against its safety requirements, many derived during a systematic safety analysis. …

    wustl Repository record for On Achieving Acceptable Levels of Safety Risk in a Reinforcement Learning Environment (opens in a new tab)

  15. Robust Reinforcement Learning via Risk-Sensitivity

    … Reinforcement Learning (RL) systems that are generic, provide performance guarantees, and can generalize-reason-improve in complex and unknown task environments. To achieve this objective, we focus on exploring the concept of Risk-sensitivity in RL systems and its …

    maryland Repository record for Robust Reinforcement Learning via Risk-Sensitivity (opens in a new tab)

  16. The Development of Robust and Safe Reinforcement Learning Methods

    Reinforcement learning (RL) has become a prominent research area in recent years due to its ability to address decision-making problems by maximizing cumulative rewards in complex environments through interactions with the environment. Concurrently, the rapid progress in deep neural network has …

    kings Repository record for The Development of Robust and Safe Reinforcement Learning Methods (opens in a new tab) Download PDF for The Development of Robust and Safe Reinforcement Learning Methods (opens in a new tab)

  17. Safety and Security of Reinforcement Learning for Autonomous Driving

    … of autonomous driving, reinforcement learning (RL) presents a powerful paradigm: agents capable of learning to drive efficiently in unseen situations through experience. However, this promise is shadowed by a fundamental concern—how can we entrust decision-making to agents that rely on …

    waterloo Repository record for Safety and Security of Reinforcement Learning for Autonomous Driving (opens in a new tab)

  18. Multi-agent optimization and learning methods for sustainable, smart and resilient power distribution systems and microgrids

    … (MGs) with costumer privacy maintenance and safety guarantee. Above problems can be directly solved by centralized solvers, which naturally requires collection of all the necessary data, reliable communication to a control center,solution to a large-scale centralized optimization problem, and …

    iastate Repository record for Multi-agent optimization and learning methods for sustainable, smart and resilient power distribution systems and microgrids (opens in a new tab)

  19. A formal methods approach to interpretability, safety and composability for reinforcement learning

    … and interpretable fashion, guarantee system safety, conveniently integrate task specific knowledge with more general knowledge about the world and generate new skills from learned ones without additional exploration. This thesis addresses these problems specifically in the case of …

    bu Repository record for A formal methods approach to interpretability, safety and composability for reinforcement learning (opens in a new tab)

  20. Responsible machine learning: supporting privacy preservation and normative alignment with multi-agent simulation

    … computational nodes geolocated around the world. It can also be inaccurate when locally simulated without concern for client parallelism, accurate timekeeping, or computation and communication loads. In this work, a recent PPFL protocol is instantiated as a single-threaded MAS to show that …

    gatech Repository record for Responsible machine learning: supporting privacy preservation and normative alignment with multi-agent simulation (opens in a new tab)