What's a Good Prediction? Challenges in evaluating an agent's knowledge

TL;DR

This paper discusses the challenges of evaluating an agent's knowledge, critiques the reliance on accuracy metrics, and proposes an alternative evaluation method based on the relevance of features to the prediction task.

Contribution

It introduces a novel evaluation approach focusing on the usefulness of predictions via feature relevance, addressing limitations of accuracy-based assessments.

Findings

Accuracy may not reflect the usefulness of knowledge.

Feature relevance provides better insight into prediction quality.

Empirical examples demonstrate the proposed evaluation method.

Abstract

Constructing general knowledge by learning task-independent models of the world can help agents solve challenging problems. However, both constructing and evaluating such models remains an open challenge. The most common approaches to evaluating models is to assess their accuracy with respect to observable values. However, the prevailing reliance on estimator accuracy as a proxy for the usefulness of the knowledge has the potential to lead us astray. We demonstrate the conflict between accuracy and usefulness through a series of illustrative examples including both a thought experiment and empirical example in MineCraft, using the General Value Function framework (GVF). Having identified challenges in assessing an agent's knowledge, we propose an alternate evaluation approach that arises continually in the online continual learning setting we recommend evaluation by examining internal learning processes, specifically the relevance of a GVF's features to the prediction task at hand. This paper contributes a first look into evaluation of predictions through their use, an integral component of predictive knowledge which is as of yet unexplored.