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.
@article{arxiv.2301.11990,
title = {Alignment with human representations supports robust few-shot learning},
author = {Ilia Sucholutsky and Thomas L. Griffiths},
journal= {arXiv preprint arXiv:2301.11990},
year = {2023}
}