Feature Learning beyond the Lazy-Rich Dichotomy: Insights from Representational Geometry

Published by

cnchou.github.io

on •1 min read

Feature Learning beyond the Lazy-Rich Dichotomy: Insights from Representational Geometry

Integrating task-relevant information into neural representations is a fundamental ability of both biological and artificial intelligence systems. Recent theories have categorized learning into two regimes: the rich regime, where neural networks actively learn task-relevant features, and the lazy...

Feature Learning Representational Geometry Neural Manifolds Manifold Capacity Lazy Regime Rich Regime Neural Tangent Kernel Deep Learning Theory ICML 2025

Learn more