Light

chandlerbang / pytorch-gnn-meta-attack Goto Github PK

Pytorch implementation of gnn meta attack (mettack). Paper title: Adversarial Attacks on Graph Neural Networks via Meta Learning.

Python 100.00%

pytorch adversarial-attacks graph-neural-networks machine-learning graph-mining deep-learning

pytorch-gnn-meta-attack's Introduction

pytorch-gnn-meta-attack

pytorch implementation of gnn meta attack (mettack). This repository is the pytorch implementation of the graph attack paper: Adversarial Attacks on Graph Neural Networks via Meta Learning

Tensorflow implementation can be found here

This method is included in DeepRobust, a very easy-to-use PyTorch Attack/Defense Library.

Requirements

Python 3.6 or newer
numpy
scipy
scikit-learn
pytorch 1.0 or newer
matplotlib (for plotting the results)
seaborn (for plotting the results)

Usage

To test the model, use the following command

python test_metattack.py

You can also add some additional configs

python test_metattack.py --dataset cora --ptb_rate 0.05 --model Meta-Self

The results on three datasets:

Cora	Citeseer	Polblogs

pytorch-gnn-meta-attack's People

Contributors

Stargazers

Watchers

Forkers

shengguanwsu pqy000 tapizz-ruixin

pytorch-gnn-meta-attack's Issues

Why use retain_graph?

Hi, I was wondering why set retain_graph=True in this line. Actually, retain_graph=True/False makes no differences to the result. Thanks.

retain_graph = True vs create_graph = True?

Hi,

In the two ** inner_train** functions in Class Metattack and MetaApprox, to calculate the gradients of weights, the parameters for the autograd.grad function are different, can you kindly explain the reasons behind this?

line 181 in metattack.py

weight_grads = torch.autograd.grad(loss_labeled, self.weights, create_graph=True)

line 327 in metattack.py

self.grad_sum += torch.autograd.grad(attack_loss, self.adj_changes, retain_graph=True)[0]

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.