Alignment with human representations supports robust few-shot learning

Ilia Sucholutsky, Thomas L. Griffiths

Keywords: representation learning, supervised learning, human alignment, few-shot learning

TL;DR: Models that are highly aligned with human representations are better at few-shot learning and are more robust to domain shift and adversarial examples

Abstract: Should we care whether AI systems have representations of the world that are similar to those of humans? We provide an information-theoretic analysis that suggests that there should be a U-shaped relationship between the degree of representational alignment with humans and performance on few-shot learning tasks. We confirm this prediction empirically, finding such a relationship in an analysis of the performance of 491 computer vision models. We also show that highly-aligned models are more robust to both natural adversarial attacks and domain shifts. Our results suggest that human-alignment is often a sufficient, but not necessary, condition for models to make effective use of limited data, be robust, and generalize well.

Submission Number: 3618