Adaptive Aggregation for Safety-Critical Control
Not on the map. Why?
Abstract (arXiv)
Safety has been recognized as the central obstacle to preventing the use of reinforcement learning (RL) for real-world applications. Different methods have been developed to deal with safety concerns in RL. However, learning reliable RL-based solutions usually require a large number of interactions with the environment. Likewise, how to improve the learning efficiency, specifically, how to utilize transfer learning for safe reinforcement learning, has not been well studied. In this work, we propose an adaptive aggregation framework for safety-critical control. Our method comprises two key techniques: 1) we learn to transfer the safety knowledge by aggregating the multiple source tasks and a target task through the attention network; 2) we separate the goal of improving task performance and reducing constraint violations by utilizing a safeguard. Experiment results demonstrate that our algorithm can achieve fewer safety violations while showing better data efficiency compared with several baselines.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 41 references that are in the map. First posted on arXiv 7 Feb 2023.
Builds on
32 of 41 references.
References · 32
- 2017Proximal Policy Optimization Algorithmscited by 14,441
- 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
- 2017Constrained Policy Optimizationcited by 928
- 2017Safe Reinforcement Learning via Shieldingcited by 427
- 2018Reward Constrained Policy Optimizationcited by 397
- 2018A Lyapunov-based Approach to Safe Reinforcement Learningcited by 278
- 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 256
- 2018Safe Exploration in Continuous Action Spacescited by 237
- 2018An Introduction to Deep Reinforcement Learningcited by 201
- 2019Deep Dynamics Models for Learning Dexterous Manipulationcited by 200
- 2020Recovery RL: Safe Reinforcement Learning with Learned Recovery Zonescited by 156
- 2020Exploration-Exploitation in Constrained MDPscited by 146
20 more
- 2017Safe Policy Improvement with Baseline Bootstrappingcited by 141
- 2020Conservative Safety Critics for Explorationcited by 107
- 2021Reinforcement Learning for Robust Parameterized Locomotion Control of Bipedal Robotscited by 95
- 2019Safe Policies for Reinforcement Learning via Primal-Dual Methodscited by 75
- 2020Safe Imitation Learning via Fast Bayesian Reward Inference from Preferencescited by 73
- 2020Safe Reinforcement Learning via Curriculum Inductioncited by 73
- 2018Safe Reinforcement Learning with Model Uncertainty Estimatescited by 70
- 2022Safe Reinforcement Learning by Imagining the Near Futurecited by 67
- 2021Safe Reinforcement Learning Using Advantage-Based Interventioncited by 45
- 2020Safety-Critical Model Predictive Control with Discrete-Time Control Barrier Functioncited by 42
- 2020Cautious Adaptation For Reinforcement Learning in Safety-Critical Settingscited by 42
- 2021Learning Barrier Certificates: Towards Safe Reinforcement Learning with Zero Training-time Violationscited by 36
- 2021Context-Aware Safe Reinforcement Learning for Non-Stationary Environmentscited by 33
- 2021Fully Autonomous Real-World Reinforcement Learning with Applications to Mobile Manipulationcited by 30
- 2020SAMBA: Safe Model-Based & Active Reinforcement Learningcited by 28
- 2021Multi-Objective SPIBB: Seldonian Offline Policy Improvement with Safety Constraints in Finite MDPscited by 22
- 2015Attend, Adapt and Transfer: Attentive Deep Architecture for Adaptive Transfer from multiple sources in the same domaincited by 19
- 2019MULTIPOLAR: Multi-Source Policy Aggregation for Transfer Reinforcement Learning between Diverse Environmental Dynamicscited by 18
- 2021Reinforcement Learning based Disease Progression Model for Alzheimer's Diseasecited by 10
- 2021Safe Policy Optimization with Local Generalized Linear Function Approximationscited by 10
Built on by
Not cited yet.