Adversarial Capability

Adversarial capability research explores how to create and defend against malicious inputs designed to deceive machine learning models. Current efforts focus on developing more effective attack methods, including those employing reinforcement learning, disentangled feature spaces, and techniques tailored to specific data types like tabular data and hardware power traces. This research is crucial for improving the robustness and security of machine learning systems across various applications, from malware detection to safety-critical systems, by identifying vulnerabilities and developing more resilient models. The ultimate goal is to create models that are not only accurate but also resistant to manipulation.

Papers

June 25, 2024

CuDA2: An approach for Incorporating Traitor Agents into Cooperative Multi-Agent Systems
Zhen Chen, Yong Liao, Youpeng Zhao, Zipeng Dai, Jian Zhao
Adversarial Attack Constructive Approach Cooperative Multi Agent Reinforcement Learning Malicious Agent Cooperative Multi Agent System Adversarial Markov Decision Process CUDA Implementation Realistic Adversarial Adversarial Capability

June 20, 2024

Uniform Convergence of Adversarially Robust Classifiers
Rachel Morris, Ryan Murray
Adversarial Perturbation Adversarial Evaluation Optimal Classifier Uniform Convergence Adversarial Capability

June 5, 2024

DifAttack++: Query-Efficient Black-Box Adversarial Attack via Hierarchical Disentangled Feature Space in Cross-Domain
Jun Liu, Jiantao Zhou, Jiandian Zeng, Jinyu Tian, Zheng Li
Adversarial Attack Adversarial Example Cross Domain Score Based Adversarial Feature Adversarial Capability

February 29, 2024

How to Train your Antivirus: RL-based Hardening through the Problem-Space
Ilias Tsingenopoulos, Jacopo Cortellazzi, Branislav Bošanský, Simone Aonzo, Davy Preuveneers, Wouter Joosen, Fabio Pierazzi, Lorenzo Cavallaro
Adversarial Example Adversarial Training Adversarial Malware Adversarial Capability Based Malware Detection

February 14, 2024

Exploring the Adversarial Capabilities of Large Language Models
Lukas Struppek, Minh Hieu Le, Dominik Hintersdorf, Kristian Kersting
Adversarial Example Adversarial Perturbation Adversarial Behavior Adversarial Capability

January 4, 2024

Evasive Hardware Trojan through Adversarial Power Trace
Behnam Omidi, Khaled N. Khasawneh, Ihsen Alouani
Side Channel Hardware Trojan Adversarial Capability Hardware Security

October 25, 2023

RAEDiff: Denoising Diffusion Probabilistic Models Based Reversible Adversarial Examples Self-Generation and Self-Recovery
Fan Xing, Xiaoyi Zhou, Xuefeng Fan, Zhuo Tian, Yan Zhao
Deep Neural Network Adversarial Example Adversarial Perturbation Denoising Diffusion Probabilistic Model Adversarial Capability Malicious Training

September 26, 2023

DifAttack: Query-Efficient Black-Box Attack via Disentangled Feature Space
Liu Jun, Zhou Jiantao, Zeng Jiandian, Jinyu Tian
Adversarial Attack Adversarial Example Feature Space Black Box Attack Score Based Adversarial Feature Adversarial Capability

August 27, 2022

Adversarial Robustness for Tabular Data through Cost and Utility Awareness
Klim Kireev, Bogdan Kulynych, Carmela Troncoso
Adversarial Example Adversarial Robustness Tabular Data Task Utility Hidden CoST Adaptive Adversary Adversarial Capability

July 21, 2022

Careful What You Wish For: on the Extraction of Adversarially Trained Models
Kacem Khaled, Gabriela Nicolescu, Felipe Gohring de Magalhães
Adversarial Example Full Model Adversarial Training Process Extraction Model Privacy Extraction Attack Adversarial Capability

May 20, 2022

Exploring the Trade-off between Plausibility, Change Intensity and Adversarial Power in Counterfactual Explanations using Multi-objective Optimization
Javier Del Ser, Alejandro Barredo-Arrieta, Natalia Díaz-Rodríguez, Francisco Herrera, Andreas Holzinger
Counterfactual Explanation Multi Objective Optimization Counterfactual Generation Counterfactual Example Explanation Plausibility Adversarial Instance Adversarial Capability