An ASR (Automatic Speech Recognition) adversarial attack repository.
-
Updated
Oct 14, 2026 - Python
An ASR (Automatic Speech Recognition) adversarial attack repository.
vanilla training and adversarial training in PyTorch
[ICLR 2026] Fine-Grained Iterative Adversarial Attacks with Limited Computation Budget — Spiking-PGD: stronger attacks & cheaper adversarial training under a fixed compute budget
Adversarial robustness evaluation of ML-based intrusion detection for IoT networks. FGSM, PGD, C&W attacks with adversarial training and ensemble defenses.
CNN robustness evaluation under PGD adversarial attacks with accuracy and perturbation analysis.
Exploring the concept of "adversarial attacks" on deep learning models, specifically focusing on image classification using PyTorch. Implementing and demonstrating the Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD) attacks against a Convolutional Neural Network (CNN) and a Recurrent Neural Network (RNN) trained on the MNIST.
Implementations for several white-box and black-box attacks.
"Neural Computing and Applications" Published Paper (2023)
TinyIDS: a <65 KB adversarially robust neural intrusion detection model for resource-constrained IoT devices. 0.85 macro-F1 on IoT-23, 0.40 ms inference, PGD adversarial training with zero inference overhead.
Evaluated the robustness of a deep face recognition model (InceptionResNetV1) against adversarial attacks. Tested multiple attack types, analyzed transferability, and implemented a defense system using specialized detectors to improve security while preserving accuracy.
Adversarial defense by retreaval-based methods
To associate your repository with the pgd-attack topic, visit your repo's landing page and select "manage topics."