Data augmentation for NLP

Advbox is a toolbox to generate adversarial examples that fool neural networks in PaddlePaddle、PyTorch、Caffe2、MxNet、Keras、TensorFlow and Advbox can benchmark the robustness of machine learning models. Advbox give a command line tool to generate adversarial examples with Zero-Coding.

A Toolbox for Adversarial Robustness Research

Pytorch implementation of convolutional neural network adversarial attack techniques

A Harder ImageNet Test Set

Tensorflow Implementation of Adversarial Attack to Capsule Networks

A non-targeted adversarial attack method, which won the first place in NIPS 2017 non-targeted adversarial attacks competition

Code for the CVPR 2019 article "Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses"

Official TensorFlow Implementation of Adversarial Training for Free! which trains robust models at no extra cost compared to natural training.

Simple pytorch implementation of FGSM and I-FGSM

PyTorch library for adversarial attack and training

A targeted adversarial attack method, which won the NIPS 2017 targeted adversarial attacks competition

Generative Adversarial Perturbations (CVPR 2018)

Deflecting Adversarial Attacks with Pixel Deflection

Wasserstein Introspective Neural Networks (CVPR 2018 Oral)

Spatially Transformed Adversarial Examples with TensorFlow

Project page for our paper: Interpreting Adversarially Trained Convolutional Neural Networks

[ICML 2019] ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation

TensorFlow version of SqueezeNet with converted pretrained weights

A simple implement of an Adversarial Autoencoding ATN(AAE ATN)

My entry for ICLR 2018 Reproducibility Challenge for paper Synthesizing robust adversarial examples https://openreview.net/pdf?id=BJDH5M-AW

reproduction of Thermometer Encoding: One Hot Way To Resist Adversarial Examples in pytorch

code for ACL 2018 paper by Kang et al., "AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples "

基于深度学习的验证码破解、对抗与反对抗研究

Adversarial Examples on KNN (and its neural network friends)

Tensorflow Implementation of adversarial learning based adversarial example generator

Capsule networks can defend against adversarial attacks using reconstruction error

An open-source textual adversarial attack toolkit.

This is a toolbox to construct adversarial examples of EEG signals. The traditional EEG extraction methods and classifiers are re-implemented in Tensorflow.

Tensorflow implementation for generating adversarial examples using convex programming