Information Directed Reward Learning for Reinforcement Learning

Information Directed Reward Learning for Reinforcement Learning

Abstract

For many reinforcement learning (RL) applications, specifying a reward is difficult. In this paper, we consider an RL setting where the agent can obtain information about the reward only by querying an expert that can, for example, evaluate individual states or provide binary preferences over trajectories. From such expensive feedback, we aim to learn a model of the reward function that allows standard RL algorithms to achieve high expected return with as few expert queries as possible. For this purpose, we propose Information Directed Reward Learning (IDRL), which uses a Bayesian model of the reward function and selects queries that maximize the information gain about the difference in return between potentially optimal policies. In contrast to prior active reward learning methods designed for specific types of queries, IDRL naturally accommodates different query types. Moreover, by shifting the focus from reducing the reward approximation error to improving the policy induced by the reward model, it achieves similar or better performance with significantly fewer queries. We support our findings with extensive evaluations in multiple environments and with different types of queries.

Grafik Top
Authors
  • Lindner, David
  • Turchetta, Matteo
  • Tschiatschek, Sebastian
  • Ciosek, Kamil
  • Krause, Andreas
Grafik Top
Shortfacts
Category
Technical Report (Working Paper)
Divisions
Data Mining and Machine Learning
Publisher
CoRR arXiv
Date
24 February 2021
Official URL
https://arxiv.org/abs/2102.12466
Export
Grafik Top