Off-policy Deep Reinforcement Learning Without Exploration

Abstract

Many practical applications of reinforcement learning constrain agents to learn from a fixed batch of data which has already been gathered, without offering further possibility for data collection. In this paper, we demonstrate that due to errors introduced by extrapolation, standard off-policy deep reinforcement learning algorithms, such as DQN and DDPG, are incapable of learning with data uncorrelated to the distribution under the current policy, making them ineffective for this fixed batch setting. We introduce a novel class of off-policy algorithms, batch-constrained reinforcement learning, which restricts the action space in order to force the agent towards behaving close to on-policy with respect to a subset of the given data. We present the first continuous control deep reinforcement learning algorithm which can learn effectively from arbitrary, fixed batch data, and empirically demonstrate the quality of its behavior in several tasks.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyfujimoto2018off

Related papers

• Off-policy Policy Gradient Algorithms By Constraining The State Distribution Shift (2019)0.00
• Handling Cost And Constraints With Off-policy Deep Reinforcement Learning (2023)0.00
• Deep Reinforcement Learning With Feedback-based Exploration (2019)5.84
• Iterative Batch Reinforcement Learning Via Safe Diversified Model-based Policy Search (2024)0.00
• Decoupled Exploration And Exploitation Policies For Sample-efficient Reinforcement Learning (2021)0.00
• Asynchronous Episodic Deep Deterministic Policy Gradient: Towards Continuous Control In Computationally Complex Environments (2019)0.00
• Off-policy Reinforcement Learning With Optimistic Exploration And Distribution Correction (2021)0.00
• Guided Exploration In Reinforcement Learning Via Monte Carlo Critic Optimization (2022)0.00