Saxe et al. (2019)

A mathematical theory of semantic development in deep neural networks

Proceedings of the National Academy of Sciences, 116(23), 11537-11546

Find It

DOI

Abstract

Derives exact solutions for the learning dynamics of deep linear networks and shows that a single simple model reproduces a range of regularities in human semantic development, including the acquisition of broad category distinctions before fine ones, transient false beliefs between those transitions, graded item typicality and category coherence, the narrowing of inductive generalisation over development, and the preservation of representational similarity structure across networks trained in the same environment.

Reference

Saxe, A. M., McClelland, J. L., & Ganguli, S. (2019). A mathematical theory of semantic development in deep neural networks. Proceedings of the National Academy of Sciences, 116(23), 11537–11546. https://doi.org/10.1073/pnas.1820226116

What Rests on This

Cited in the argument rather than behind a score: this is one of the papers the taxonomy is built on, named where the site explains itself rather than in the findings under a mechanism.