Uncertainty Weighted Actor-Critic for Offline Reinforcement Learning

2021-05-17 · Yue Wu, Shuangfei Zhai, Nitish Srivastava, Joshua Susskind, Jian Zhang, Ruslan Salakhutdinov, Hanlin Goh

Offline Reinforcement Learning promises to learn effective policies from previously-collected, static datasets without the need for exploration. However, existing Q-learning and actor-critic based off-policy RL algorithms fail when bootstrapping from out-of-distribution (OOD) actions or states. We hypothesize that a key missing ingredient from the existing methods is a proper treatment of uncertainty in the offline setting. We propose Uncertainty Weighted Actor-Critic (UWAC), an algorithm that detects OOD state-action pairs and down-weights their contribution in the training objectives accordingly. Implementation-wise, we adopt a practical and effective dropout-based uncertainty estimation method that introduces very little overhead over existing RL algorithms. Empirically, we observe that UWAC substantially improves model stability during training. In addition, UWAC out-performs existing offline RL methods on a variety of competitive tasks, and achieves significant performance gains over the state-of-the-art baseline on datasets with sparse demonstrations collected from human experts.

📄 PDF Abstract BibTeX arXiv:2105.08140

Code (2)

Holmeswww/ml-uwac 공식 구현 pytorch

apple/ml-uwac 공식 구현 pytorch

Tasks

Offline RLQ-Learningreinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

Q-Learning Q-Learning is an off-policy temporal difference control algorithm: $$Q\left(S\_{t}, A\_{t}\right) \leftarrow Q\left(S\_{t}, A\_{t}\right) + \alpha\left[R_{t+1} +…

Uncertainty Weighted Offline Reinforcement Learning

2021-01-01 · Yue Wu, Shuangfei Zhai, Nitish Srivastava, Joshua M. Susskind 외

Offline RL Q-Learning reinforcement-learning Reinforcement Learning+1

Importance Weighted Actor-Critic for Optimal Conservative Offline Reinforcement Learning

2023-01-30 · NeurIPS 2023 11 · Hanlin Zhu, Paria Rashidinejad, Jiantao Jiao

We propose A-Crab (Actor-Critic Regularized by Average Bellman error), a new practical algorithm for offline reinforcement learning (RL) in complex environments with insufficient data coverage. Our algorithm combines the…

reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Adversarially Trained Weighted Actor-Critic for Safe Offline Reinforcement Learning

2024-01-01 · Honghao Wei, Xiyue Peng, Arnob Ghosh, Xin Liu

We propose WSAC (Weighted Safe Actor-Critic), a novel algorithm for Safe Offline Reinforcement Learning (RL) under functional approximation, which can robustly optimize policies to improve upon an arbitrary reference pol…

continuous-control Continuous Control Offline RL Reinforcement Learning (RL)

Q-Value Weighted Regression: Reinforcement Learning with Limited Data

2021-02-12 · Piotr Kozakowski, Łukasz Kaiser, Henryk Michalewski, Afroz Mohiuddin 외

Sample efficiency and performance in the offline setting have emerged as significant challenges of deep reinforcement learning. We introduce Q-Value Weighted Regression (QWR), a simple RL algorithm that excels in these a…

Atari Games continuous-control Continuous Control Deep Reinforcement Learning+6

FAWAC: Feasibility Informed Advantage Weighted Regression for Persistent Safety in Offline Reinforcement Learning

2024-12-12 · Prajwal Koirala, Zhanhong Jiang, Soumik Sarkar, Cody Fleming

Safe offline reinforcement learning aims to learn policies that maximize cumulative rewards while adhering to safety constraints, using only offline data for training. A key challenge is balancing safety and performance,…