CIER: A Novel Experience Replay Approach with Causal Inference in Deep Reinforcement Learning
This addresses data efficiency and interpretability issues in DRL training, offering an incremental improvement over existing methods.
The paper tackles the challenge of enhancing data utilization and explainability in Deep Reinforcement Learning (DRL) by proposing CIER, a novel experience replay approach that uses causal inference on segmented time series to identify key causal factors impacting training outcomes. Experiments in common environments show it improves DRL training effectiveness and adds explainability, with continued effectiveness when extended with priority experience replay.
In the training process of Deep Reinforcement Learning (DRL), agents require repetitive interactions with the environment. With an increase in training volume and model complexity, it is still a challenging problem to enhance data utilization and explainability of DRL training. This paper addresses these challenges by focusing on the temporal correlations within the time dimension of time series. We propose a novel approach to segment multivariate time series into meaningful subsequences and represent the time series based on these subsequences. Furthermore, the subsequences are employed for causal inference to identify fundamental causal factors that significantly impact training outcomes. We design a module to provide feedback on the causality during DRL training. Several experiments demonstrate the feasibility of our approach in common environments, confirming its ability to enhance the effectiveness of DRL training and impart a certain level of explainability to the training process. Additionally, we extended our approach with priority experience replay algorithm, and experimental results demonstrate the continued effectiveness of our approach.