Skip to content

Latest commit

 

History

History
28 lines (19 loc) · 1.89 KB

AWESOME_LIST.md

File metadata and controls

28 lines (19 loc) · 1.89 KB

Awesome List

There is a lot of awesome research and development happening out in the interpretability community that we would like to share. Here we will maintain a curated list of research, implementations and resources. We would love to learn about more! Please feel free to make a pull request to contribute to the list.

TorchRay: Visualization methods for deep CNNs

TorchRay focuses on attribution, namely the problem of determining which part of the input, usually an image, is responsible for the value computed by a neural network.

Score Cam: A gradient-free CAM extension

Score-CAM is a gradient-free visualization method extended from Grad-CAM and Grad-CAM++. It provides score-weighted visual explanations for CNNs.

White Noise Analysis

White noise stimuli is fed to a classifier and the ones that are categorized into a particular class are averaged. It gives an estimate of the templates a classifier uses for classification, and is based on two popular and related methods in psychophysics and neurophysiology namely classification images and spike triggered analysis.

FastCAM: Multiscale Saliency Map with SMOE scale

An attribution method that uses information at the end of each network scale which is then combined into a single saliency map.