An interesting pattern in the final fully connected layer weights

TL;DR I have written a more formal summary of my findings in here. Github: https://github.com/sidml/interpret-fc-layer You can replicate my experiment results by running the notebooks in git repo or by running this Kaggle Notebook for fitting ALD and this for visualizing the internal split. The results of fitting ALD to FC weights are here and visualization of the most discriminative neurons using Smooth GramCam++ is here. Motivation Like everyone else, I have struggled to interpret the pre...

Submit