adverml Goto Github PK

lid_adversarial_subspace_detection

Code for paper "Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality".

machine_learning_deep_knn

Experiments with deep learning interpretability based on "Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning" (https://arxiv.org/abs/1803.04765)

multilid

nnif_adv_defense

Detection of adversarial examples using influence functions and nearest neighbors

novelty-detection

Analyzing basic network responses to novel classes

patch-fool

[ICLR 2022] "Patch-Fool: Are Vision Transformers Always Robust Against Adversarial Perturbations?" by Yonggan Fu, Shunyao Zhang, Shang Wu, Cheng Wan, Yingyan Lin

perceptual-advex

Code and data for the ICLR 2021 paper "Perceptual Adversarial Robustness: Defense Against Unseen Threat Models".

pytorch

pytorch-classification

Classification with PyTorch.

pytorch-ddpm-cifar10

Unofficial PyTorch implementation of Denoising Diffusion Probabilistic Models

robust-ood-detection

Robust Out-of-distribution Detection in Neural Networks

robust_classification_with_detection

Source code for Sheikholeslami et al., "Provably Robust Classification of Adversarial Examples with Detection", ICLR 2021.

robustness

A library for experimenting with, training and evaluating neural networks, with a focus on adversarial robustness.

sammd

This is the source code for Maximum Mean Discrepancy Test is Aware of Adversarial Attacks (ICML2021).

semisup-adv

Semisupervised learning for adversarial robustness https://arxiv.org/pdf/1905.13736.pdf

spectraladversarialdefense

spectraldef_framework

Is RobustBench/AutoAttack a suitable Benchmark for Adversarial Robustness?

spectraldefense

synthesizing_robust_adversarial

PyTorch implementation of Expectation over Transformation

tf_to_pytorch_model

Convert tensorflow model to pytorch model via [MMdnn](https://github.com/microsoft/MMdnn) for adversarial attacks.

trades

TRADES (TRadeoff-inspired Adversarial DEfense via Surrogate-loss minimization)

trustscore

To Trust Or Not To Trust A Classifier. A measure of uncertainty for any trained (possibly black-box) classifier which is more effective than the classifier's own implied confidence (e.g. softmax probability for a neural network).

vim

Official repository for CVPR2022 publication, ViM: Out-Of-Distribution with Virtual-logit Matching

visual_prompting

Exploring Visual Prompts for Adapting Large-Scale Models

vp-for-adversarial-robustness

Visual Prompting for Adversarial Robustness