Gradient Attack

Gradient attacks exploit the gradients of machine learning models to generate adversarial inputs—data subtly modified to cause misclassification or data leakage. Current research focuses on improving the effectiveness and transferability of these attacks across various model architectures, including convolutional neural networks and transformers, and in different contexts like federated learning and graph neural networks. This area is crucial for assessing the robustness of machine learning systems and for developing effective defenses against privacy violations and model manipulation, impacting the security and reliability of AI applications.

Papers

November 6, 2024

Optimal Defenses Against Gradient Reconstruction Attacks
Yuxiao Chen, Gamze Gürsoy, Qi Lei
Data Leakage Defense Algorithm Gradient Noise Gradient Attack Gradient Pruning

October 28, 2024

Inverting Gradient Attacks Naturally Makes Data Poisons: An Availability Attack on Neural Networks
Wassim Bouaziz, El-Mahdi El-Mhamdi, Nicolas Usunier
Neural Network Data Poisoning Gradient Inversion Attack Gradient Based Attack Resource Utilization Gradient Attack Availability Attack

October 21, 2024

Extracting Spatiotemporal Data from Gradients with Large Language Models
Lele Zheng, Yang Cao, Renhe Jiang, Kenjiro Taura, Yulong Shen, Sheng Li, Masatoshi Yoshikawa
Spatio tempOral Gradient Inversion Attack SpatioTemporal Data Gradient Attack

October 4, 2024

Gradient-based Jailbreak Images for Multimodal Fusion Models
Javier Rando, Hannah Korevaar, Erik Brinkman, Ivan Evtimov, Florian Tramèr
Jailbreak Attack Adversarial Input Text Attack Multimodal Fusion Model Gradient Attack Representation Engineering

September 26, 2024

Federated Learning under Attack: Improving Gradient Inversion for Batch of Images
Luiz Leite, Yuri Santo, Bruno L. Dalmazo, André Riker
Attack Success Rate Gradient Inversion Attack Gradient Inversion Gradient Attack Batch Greenkhorn

July 9, 2024

Improving the Transferability of Adversarial Examples by Feature Augmentation
Donghua Wang, Wen Yao, Tingsong Jiang, Xiaohu Zheng, Junqi Wu, Xiaoqian Chen
Adversarial Example Task Transferability Feature Augmentation Gradient Attack

June 19, 2024

AGSOA:Graph Neural Network Targeted Attack Based on Average Gradient and Structure Optimization
Yang Chen, Bin Zhou
Adversarial Attack Gradient Information Gradient Attack Continuous Gradient Structural Optimization

June 6, 2024

R-CONV: An Analytical Approach for Efficient Data Reconstruction via Convolutional Gradients
Tamer Ahmed Eltaras, Qutaibah Malluhi, Alessandro Savino, Stefano Di Carlo, Adnan Qayyum, Junaid Qadir
Convolutional Layer Convolution Layer Gradient Attack Analytical Approach Data Reconstruction Gradient Sharing

June 2, 2024

Constrained Adaptive Attack: Effective Adversarial Attack Against Deep Neural Networks for Tabular Data
Thibault Simonetto, Salah Ghamizi, Maxime Cordy
Deep Neural Network Adversarial Robustness Tabular Data Adaptive Attack Tabular Deep Learning Gradient Attack Effective Adversarial

February 26, 2024

A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection
Leonid Boytsov, Ameya Joshi, Filipe Condessa
Tiny Refinement Elicit Resilience Gradient Based Attack Closed Form Differentiable Expression Gradient Attack Gradient Masking

February 12, 2024

Correctness Verification of Neural Networks Approximating Differential Equations
Petros Ellinas, Rahul Nellikath, Ignasi Ventura, Jochen Stiasny, Spyros Chatzivasileiadis
Differential Equation Finite Difference Domain Constraint Gradient Attack Image Derivative Correctness Check Residual Bounding

January 30, 2024

Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks
Lulu Xue, Shengshan Hu, Ruizhi Zhao, Leo Yu Zhang, Shengqing Hu, Lichao Sun, Dezhong Yao
Differential Privacy Gradient Inversion Attack Defense Method Gradient Attack Perturbation Defense Gradient Pruning

November 22, 2023

A Theoretical Insight into Attack and Defense of Gradient Leakage in Transformer
Chenyang Li, Zhao Song, Weixin Wang, Chiwun Yang
Transformer Based Differential Privacy Theoretical Analysis Gradient Leakage Gradient Attack Gradient Leakage Attack

March 12, 2023

Adaptive Local Adversarial Attacks on 3D Point Clouds for Augmented Reality
Weiquan Liu, Shijun Zheng, Cheng Wang
Adversarial Example 3D Point Cloud Augmented Reality Gradient Attack Adversarial Point Cloud 3D Adversarial Adaptive Adversarial Attack

October 28, 2022

Secure Distributed Optimization Under Gradient Attacks
Shuhua Yu, Soummya Kar
Optimization Purpose Stochastic Gradient Secure Approach Multi Agent Network Gradient Attack Gradient Oracle

August 9, 2022

Combining Variational Modeling with Partial Gradient Perturbation to Prevent Deep Gradient Leakage
Daniel Scheliga, Patrick Mäder, Marco Seeland
Gradient Inversion Attack Variational Model Gradient Leakage Gradient Attack Gradient Perturbation

February 17, 2022

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks
Sadegh Farhadkhani, Rachid Guerraoui, Lê-Nguyên Hoang, Oscar Villemaud
Data Poisoning Gradient Attack Impossibility Theorem Byzantine Machine Learning

January 18, 2022

How to Backdoor HyperNetwork in Personalized Federated Learning?
Phung Lai, NhatHai Phan, Issa Khalil, Abdallah Khreishah, Xintao Wu
Personalized Federated Learning Non Graph Hypernetwork Baseline Model Stealing Attack Model Attack Gradient Attack

December 23, 2021

Adaptive Modeling Against Adversarial Attacks
Zhiwen Yan, Teck Khim Ng
Adversarial Attack Adversarial Training Adversarial Input Adversarial Data Adversarial Defense Adaptive Model Gradient Attack

Gradient Attack

Papers

Optimal Defenses Against Gradient Reconstruction Attacks

Inverting Gradient Attacks Naturally Makes Data Poisons: An Availability Attack on Neural Networks

Extracting Spatiotemporal Data from Gradients with Large Language Models

Gradient-based Jailbreak Images for Multimodal Fusion Models

Federated Learning under Attack: Improving Gradient Inversion for Batch of Images

Improving the Transferability of Adversarial Examples by Feature Augmentation

AGSOA:Graph Neural Network Targeted Attack Based on Average Gradient and Structure Optimization

R-CONV: An Analytical Approach for Efficient Data Reconstruction via Convolutional Gradients

Constrained Adaptive Attack: Effective Adversarial Attack Against Deep Neural Networks for Tabular Data

A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection

Correctness Verification of Neural Networks Approximating Differential Equations

Revisiting Gradient Pruning: A Dual Realization for Defending against Gradient Attacks

A Theoretical Insight into Attack and Defense of Gradient Leakage in Transformer

Adaptive Local Adversarial Attacks on 3D Point Clouds for Augmented Reality

Secure Distributed Optimization Under Gradient Attacks

Combining Variational Modeling with Partial Gradient Perturbation to Prevent Deep Gradient Leakage

An Equivalence Between Data Poisoning and Byzantine Gradient Attacks

How to Backdoor HyperNetwork in Personalized Federated Learning?

Adaptive Modeling Against Adversarial Attacks