Adversarial Retraining

Adversarial retraining aims to improve the robustness of machine learning models against adversarial attacks, where malicious inputs are designed to fool the model. Current research focuses on developing more resilient adversarial detectors, adapting models efficiently to new domains (e.g., using reprogramming techniques), and creating robust defense mechanisms, such as double defense strategies combining novelty detection with retraining. This work is crucial for enhancing the security and reliability of machine learning systems across various applications, from industrial IoT analytics to natural language processing, by mitigating the impact of adversarial examples.

Papers

October 14, 2024

ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection
Martin Aubard, László Antal, Ana Madureira, Luis F. Teixeira, Erika Ábrahám
Autonomous Underwater Vehicle Adversarial Patch Attack Side Scan Sonar Adversarial Datasets Adversarial Retraining

June 19, 2024

Jogging the Memory of Unlearned LLMs Through Targeted Relearning Attacks
Shengyuan Hu, Yiwei Fu, Zhiwei Steven Wu, Virginia Smith
Machine Unlearning Memory Trace Unlearning Framework Harmful Unlearning Unlearned Model Adversarial Retraining

April 18, 2024

Fortify the Guardian, Not the Treasure: Resilient Adversarial Detectors
Raz Lapid, Almog Dubin, Moshe Sipper
Adversarial Detection Treasure Trove Adversarial Retraining Adaptive Adversarial Attack

September 24, 2023

Efficient Black-Box Speaker Verification Model Adaptation with Reprogramming and Backend Learning
Jingyu Li, Tan Lee
Deep Neural Network Domain Adaptation Speaker Verification Back End Adversarial Retraining Channel Wise Lightweight Reprogramming

January 23, 2023

DODEM: DOuble DEfense Mechanism Against Adversarial Attacks Towards Secure Industrial Internet of Things Analytics
Onat Gungor, Tajana Rosing, Baris Aksanli
Adversarial Attack DEfense Mechanism Adversarial Retraining

June 8, 2022

Adversarial Text Normalization
Joanna Bitton, Maya Pavlova, Ivan Evtimov
Adversarial Attack Adversarial Data Text Normalization Adversarial Retraining

June 7, 2022

Adversarial Reprogramming Revisited
Matthias Englert, Ranko Lazic
Gradient Flow Two Layer ReLU Batch Normalization Layer Untrained Neural Network Adversarial Retraining

March 31, 2022

Exploring Visual Prompts for Adapting Large-Scale Models
Hyojin Bahng, Ali Jahanian, Swami Sankaranarayanan, Phillip Isola
Pre Trained Model Visual Prompt Visual Prompting Large Scale Model AdaPtation Performance Adversarial Retraining

February 16, 2022

The Adversarial Security Mitigations of mmWave Beamforming Prediction Models using Defensive Distillation and Adversarial Retraining
Murat Kuzlu, Ferhat Ozgur Catak, Umit Cali, Evren Catak, Ozgur Guler
Deep Neural Network Adversarial Attack Adversarial Example mmWave Signal Prediction Model Beam Prediction Adversarial Retraining Defensive Distillation

December 2, 2021

A Unified Framework for Adversarial Attack and Defense in Constrained Feature Space
Thibault Simonetto, Salijona Dyrmishi, Salah Ghamizi, Maxime Cordy, Yves Le Traon
Adversarial Attack Adversarial Example Adversarial Robustness Unified Framework Gradient Based Attack Adversarial Retraining

Adversarial Retraining

Papers

ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection

Jogging the Memory of Unlearned LLMs Through Targeted Relearning Attacks

Fortify the Guardian, Not the Treasure: Resilient Adversarial Detectors

Efficient Black-Box Speaker Verification Model Adaptation with Reprogramming and Backend Learning

DODEM: DOuble DEfense Mechanism Against Adversarial Attacks Towards Secure Industrial Internet of Things Analytics

Adversarial Text Normalization

Adversarial Reprogramming Revisited

Exploring Visual Prompts for Adapting Large-Scale Models

The Adversarial Security Mitigations of mmWave Beamforming Prediction Models using Defensive Distillation and Adversarial Retraining

A Unified Framework for Adversarial Attack and Defense in Constrained Feature Space