Offline RL With Resource Constrained Online Deployment

Abstract

Offline reinforcement learning is used to train policies in scenarios where real-time access to the environment is expensive or impossible. As a natural consequence of these harsh conditions, an agent may lack the resources to fully observe the online environment before taking an action. We dub this situation the resource-constrained setting. This leads to situations where the offline dataset (available for training) can contain fully processed features (using powerful language models, image models, complex sensors, etc.) which are not available when actions are actually taken online. This disconnect leads to an interesting and unexplored problem in offline RL: Is it possible to use a richly processed offline dataset to train a policy which has access to fewer features in the online environment? In this work, we introduce and formalize this novel resource-constrained problem setting. We highlight the performance gap between policies trained using the full offline dataset and policies trained using limited features. We address this performance gap with a policy transfer algorithm which first trains a teacher agent using the offline dataset where features are fully available, and then transfers this knowledge to a student agent that only uses the resource-constrained features. To better capture the challenge of this setting, we propose a data collection procedure: Resource Constrained-Datasets for RL (RC-D4RL). We evaluate our transfer algorithm on RC-D4RL and the popular D4RL benchmarks and observe consistent improvement over the baseline (TD3+BC without transfer). The code for the experiments is available at https://github.com/JayanthRR/RC-OfflineRL.

arXiv commentAdded experiments on discrete control and real world datasets along with more analyses on continuous control tasks

References in corpus (23)

• Continuous control with deep reinforcement learning
• Playing Atari with Deep Reinforcement Learning
• Knowledge Distillation: A Survey
• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
• Conservative Q-Learning for Offline Reinforcement Learning
• Decision Transformer: Reinforcement Learning via Sequence Modeling
• Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction
• Deep Reinforcement Learning that Matters
• D4RL: Datasets for Deep Data-Driven Reinforcement Learning
• Behavior Regularized Offline Reinforcement Learning
• A Minimalist Approach to Offline Reinforcement Learning
• MOReL : Model-Based Offline Reinforcement Learning
• Transfer Learning in Deep Reinforcement Learning: A Survey
• Domain Generalization by Marginal Transfer Learning
• Regret Bounds for Restless Markov Bandits
• Offline Reinforcement Learning with Fisher Divergence Critic Regularization
• The Importance of Pessimism in Fixed-Dataset Policy Optimization
• Uncertainty Weighted Actor-Critic for Offline Reinforcement Learning
• A Generalization Error Bound for Multi-class Domain Generalization
• Offline RL Without Off-Policy Evaluation
• Simple Regret Minimization for Contextual Bandits
• Behavioral Priors and Dynamics Models: Improving Performance and Domain Transfer in Offline RL