Results 51 to 60 of about 7,404,725 (288)
Safe Reinforcement Learning in Uncertain Contexts
IEEE Transactions on RoboticsWhen deploying machine learning algorithms in the real world, guaranteeing safety is an essential asset. Existing safe learning approaches typically consider continuous variables, i.e., regression tasks. However, in practice, robotic systems are also subject to discrete, external environmental changes, e.g., having to carry objects of certain weights ...
Dominik Baumann, Thomas B. Schön
openaire +4 more sources
Advances in Sustainable and Wearable Textile Based Soft Robotics
Advanced Functional Materials, EarlyView.This Review examines advances in wearable textile‐based soft robotics, focusing on sustainable materials, integrated sensing, and scalable actuation. It discusses manufacturing and system integration across healthcare, assistive robotics, prosthetics, and human–machine interfaces, and highlights key challenges in circular design, including life‐cycle ...
Zahir Abbas +6 more
wiley +1 more source
PILCO: A Model-Based and Data-Efficient Approach to Policy Search [PDF]
, 2011 04.07.13 KB. Ok to add accepted version to Spiral. Authors retain copyright.In this paper, we introduce PILCO, a practical, data-efficient model-based policy search method.
Rasmussen, Carl E, Deisenroth, Marc P
core
Safe Reinforcement Learning With Dual Robustness
IEEE Transactions on Pattern Analysis and Machine IntelligenceReinforcement learning (RL) agents are vulnerable to adversarial disturbances, which can deteriorate task performance or compromise safety specifications. Existing methods either address safety requirements under the assumption of no adversary (e.g., safe RL) or only focus on robustness against performance adversaries (e.g., robust RL).
Zeyang Li 0001 +4 more
openaire +4 more sources
Safe Reinforcement Learning via Shielding
Proceedings of the AAAI Conference on Artificial Intelligence, 2018 Reinforcement learning algorithms discover policies that maximize reward, but do not necessarily guarantee safety during learning or execution phases. We introduce a new approach to learn optimal policies while enforcing properties expressed in temporal logic.
Mohammed Alshiekh +5 more
openaire +2 more sources
Advanced Functional Materials, EarlyView.Advanced ink systems for solution‐processed textile triboelectric nanogenerators are systematically summarized, spanning conductive, tribo‐negative, and tribo‐positive layers. By connecting ink chemistry, deposition methods, and device function, the present review reveals the key governing principles of solution development and highlights practical ...
Xinlong Sun, Stephen Beeby
wiley +1 more source
Refined Risk Management in Safe Reinforcement Learning with a Distributional Safety Critic
, 2022 Safety is critical to broadening the real-world use of reinforcement learning (RL). Modeling the safety aspects using a safety-cost signal separate from the reward is becoming standard practice, since it avoids the problem of finding a good balance ...
Yang, Q. (author) +3 more
core
Revisiting Safe Exploration in Safe Reinforcement learning
CoRRSafe reinforcement learning (SafeRL) extends standard reinforcement learning with the idea of safety, where safety is typically defined through the constraint of the expected cost return of a trajectory being below a set limit. However, this metric fails to distinguish how costs accrue, treating infrequent severe cost events as equal to frequent mild ...
David Eckel +2 more
openaire +2 more sources
Safe RLHF: Safe Reinforcement Learning from Human Feedback
CoRR, 2023 With the development of large language models (LLMs), striking a balance between the performance and safety of AI systems has never been more critical. However, the inherent tension between the objectives of helpfulness and harmlessness presents a significant challenge during LLM training.
Josef Dai +7 more
openaire +4 more sources
Leveraging Analytic Gradients in Provably Safe Reinforcement Learning
IEEE Open Journal of Control SystemsThe deployment of autonomous robots in safety-critical applications requires safety guarantees. Provably safe reinforcement learning is an active field of research that aims to provide such guarantees using safeguards.
Tim Walter +3 more
doaj +1 more source