Representing expertise accelerates learning from pedagogical interaction data

TL;DR

This paper investigates how representing expertise in interaction data enhances learning efficiency in AI models, demonstrating that pedagogical interaction data improves robustness and enables expert-like behavior.

Contribution

It introduces a controlled paradigm to analyze interaction features, showing that modeling epistemic distinctions in interactions boosts learning outcomes.

Findings

Models trained on pedagogical interactions outperform those trained only on expert demonstrations.

Representing epistemic distinctions enables models to exhibit expert-like behavior with limited expert data.

Interaction data improves robustness across various scenarios.

Abstract

Work in cognitive science and artificial intelligence has suggested that exposing learning agents to traces of interaction between multiple individuals can improve performance in a variety of settings, yet it remains unknown which features of interactions contribute to this improvement. We examined the factors that support the effectiveness of interaction data, using a controlled paradigm that allowed us to precisely operationalize key distinctions between interaction and an expert acting alone. We generated synthetic datasets of simple interactions between an expert and a novice in a spatial navigation task, and then trained transformer models on those datasets, evaluating performance after exposure to different datasets. Our experiments showed that models trained on pedagogical interactions were more robust across a variety of scenarios compared to models trained only on expert demonstrations, and that having the ability to represent epistemically distinct agents led to expert-like behavior even when expert behavior was rarely observed.