Awesome

aaai_hydra

Introduction

This repository contains the implementation code of HYDRA: Hypergradient Data Relevance Analysis for Interpreting Deep Neural Networks.

HYDRA is a method of neural network interpretability that assesses the contribution of training data. You can play this demo to get a feel for its power.

Installation

Dependency

Supported Operating Systems

Linux and Windows should work.

Software dependency

PyTorch >= 2.1
A C++20 compiler to build a PyTorch extension

Install

Here it's assumed that pip is used as the package manager.

Install the dependent libraries:

pip3 install -r requirements.txt --user

Test

bash hydra_train.sh

Citation

If you find our work useful, please cite it:

@article{chen2021hydra,
  title={Hydra: Hypergradient data relevance analysis for interpreting deep neural networks},
  author={Chen, Yuanyuan and Li, Boyang and Yu, Han and Wu, Pengcheng and Miao, Chunyan},
  journal={arXiv preprint arXiv:2102.02515},
  year={2021}
}