A Generalized Acquisition Function for Preference-based Reward Learning

TL;DR

This paper introduces a new acquisition function for preference-based reward learning that focuses on learning reward functions up to behavioral equivalence, improving data efficiency and performance over existing methods.

Contribution

It proposes a generalized framework for preference query synthesis that targets reward functions up to behavioral equivalence classes, enhancing learning efficiency.

Findings

Outperforms state-of-the-art information gain methods in experiments

Effective in synthetic, robotics, and NLP environments

Demonstrates improved data efficiency and reward learning accuracy

Abstract

Preference-based reward learning is a popular technique for teaching robots and autonomous systems how a human user wants them to perform a task. Previous works have shown that actively synthesizing preference queries to maximize information gain about the reward function parameters improves data efficiency. The information gain criterion focuses on precisely identifying all parameters of the reward function. This can potentially be wasteful as many parameters may result in the same reward, and many rewards may result in the same behavior in the downstream tasks. Instead, we show that it is possible to optimize for learning the reward function up to a behavioral equivalence class, such as inducing the same ranking over behaviors, distribution over choices, or other related definitions of what makes two rewards similar. We introduce a tractable framework that can capture such definitions of similarity. Our experiments in a synthetic environment, an assistive robotics environment with domain transfer, and a natural language processing problem with real datasets demonstrate the superior performance of our querying method over the state-of-the-art information gain method.