PsiPhi-Learning: Reinforcement Learning with Demonstrations using Successor Features and Inverse Temporal Difference Learning

TL;DR

PsiPhi-Learning introduces a novel reinforcement learning framework that leverages successor features and inverse temporal difference learning to utilize demonstrations without reward labels, enhancing transfer and imitation capabilities.

Contribution

It proposes a new multi-task IRL algorithm, ITD, that learns shared features and agent-specific preferences from demonstrations without reward information, integrated into RL with demonstrations.

Findings

Effective in improving RL, IRL, and imitation performance

Enables few-shot transfer to new tasks

Provides theoretical bounds for zero-shot transfer

Abstract

We study reinforcement learning (RL) with no-reward demonstrations, a setting in which an RL agent has access to additional data from the interaction of other agents with the same environment. However, it has no access to the rewards or goals of these agents, and their objectives and levels of expertise may vary widely. These assumptions are common in multi-agent settings, such as autonomous driving. To effectively use this data, we turn to the framework of successor features. This allows us to disentangle shared features and dynamics of the environment from agent-specific rewards and policies. We propose a multi-task inverse reinforcement learning (IRL) algorithm, called \emph{inverse temporal difference learning} (ITD), that learns shared state features, alongside per-agent successor features and preference vectors, purely from demonstrations without reward labels. We further show how to seamlessly integrate ITD with learning from online environment interactions, arriving at a novel algorithm for reinforcement learning with demonstrations, called $\Psi \Phi$-learning (pronounced `Sci-Fi'). We provide empirical evidence for the effectiveness of $\Psi \Phi$-learning as a method for improving RL, IRL, imitation, and few-shot transfer, and derive worst-case bounds for its performance in zero-shot transfer to new tasks.