Popular GitHub Repositories for Pgd

Pytorch-Named-Entity-Recognition-with-transformers

Implementation of gradient-based adversarial attack(FGSM,MI-FGSM,PGD)

Implementation of adversarial training under fast-gradient sign method (FGSM), projected gradient descent (PGD) and CW using Wide-ResNet-28-10 on cifar-10. Sample code is re-usable despite changing the model or dataset.

An ASR (Automatic Speech Recognition) adversarial attack repository.

Reproduce multiple adversarial attack methods

Official implementation of the paper: "REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions", IEEE WACV, 2022

Evaluating CNN robustness against various adversarial attacks, including FGSM and PGD.

This work is based on enhancing the robustness of targeted classifier models against adversarial attacks. To achieve this, a convolutional autoencoder-based approach is employed that effectively counters adversarial perturbations introduced to the input images.

Implementation of PGD attack on a model trained on cifar10 dataset in TensorFlow. Also, FID between original images and generated images has been calculated.

In the dynamic landscape of medical artificial intelligence, this study explores the vulnerabilities of the Pathology Language-Image Pretraining (PLIP) model, a Vision Language Foundation model, under targeted attacks like PGD adversarial attack.

Code implementation of our AAAI'22 paper "Improved Gradient-Based Adversarial Attacks for Quantized Networks"

WideResNet implementation on MNIST dataset. FGSM and PGD adversarial attacks on standard training, PGD adversarial training, and Feature Scattering adversarial training.

A classical or convolutional neural network model with adversarial defense protection

Projected Gradient Descent (PGD), inverted and amplified -> prompt & generate images with CLIP

Adversarial Network Attacks (PGD, pixel, FGSM) Noise on MNIST Images Dataset using Python (Pytorch)

my MA thesis (code, paper & presentation) about adversarial out-of-distribution detection

An University Project for the AI4Cybersecurity class.

2022 Spring Semester, Personal Project Research

An implementation of a deepfake detection model that uses gradient regularization to improve robustness against adversarial attacks. This approach perturbs the mean and standard deviation of shallow layers in an EfficientNetB0 backbone to enhance generalization and defend against attacks like FGSM and PGD.

Explore adversarial attacks on traffic sign recognition models and evaluate defenses using adversarial training. Includes FGSM, PGD, BIM attacks, and robust model comparison through an interactive Streamlit app.