Yizhou Xu

Scaling Laws and Spectra of Shallow Neural Networks in the Feature Learning Regime

Information-Theoretic Criteria for Knowledge Distillation in Multimodal Learning

Neural Thermodynamics: Entropic Forces in Deep and Universal Representation Learning

Learning with Restricted Boltzmann Machines: Asymptotics of AMP and GD in High Dimensions

Three Mechanisms of Feature Learning in a Linear Network

Remove Symmetries to Control Model Expressivity and Improve Optimization