Policy Gradients Incorporating the Future
arXiv:2108.02096
Abstract
Reasoning about the future -- understanding how decisions in the present time affect outcomes in the future -- is one of the central challenges for reinforcement learning (RL), especially in highly-stochastic or partially observable environments. While predicting the future directly is hard, in this work we introduce a method that allows an agent to "look into the future" without explicitly predicting it. Namely, we propose to allow an agent, during its training on past experience, to observe what \emph{actually} happened in the future at that time, while enforcing an information bottleneck to avoid the agent overly relying on this privileged information. This gives our agent the opportunity to utilize rich and useful information about the future trajectory dynamics in addition to the present. Our method, Policy Gradients Incorporating the Future (PGIF), is easy to implement and versatile, being applicable to virtually any policy gradient algorithm. We apply our proposed method to a number of off-the-shelf RL algorithms and show that PGIF is able to achieve higher reward faster in a variety of online and offline RL domains, as well as sparse-reward and partially observable environments.
References in corpus (8)
- Professor Forcing: A New Algorithm for Training Recurrent Networks
- Benchmarking Model-Based Reinforcement Learning
- Learning Stochastic Recurrent Networks
- The Optimal Reward Baseline for Gradient-Based Reinforcement Learning
- AlgaeDICE: Policy Gradient from Arbitrary Experience
- Learning model-based planning from scratch
- Representation Matters: Offline Pretraining for Sequential Decision Making
- Hindsight Credit Assignment