Weight Interpretability

Creator
Creator
Seonglae Cho
Created
Created
2024 Nov 18 22:24
Editor
Edited
Edited
2025 Jun 13 18:11

Parameter Interpretability

Weights are a vector in parameter space. Attribution is an effect of weight and feature is an effect of representation. The motivation for the weight-similarity is to avoid components sharing param.
Weight Interpretability Notion
 
 
Weight Interpretability Methods
 
 
 
Bilinear MLPs
Achille and Soatto (2018) studied the amount of information stored in the weights of deep networks
 

Recommendations