Adversary Model

Adversary models are used to evaluate the robustness and security of various systems, particularly in machine learning and distributed computing. Current research focuses on developing sophisticated adversary models that capture realistic attack strategies, including those that adapt to defensive measures, and on designing robust defenses against these attacks. This work is crucial for improving the security and reliability of systems ranging from federated learning platforms to blockchain networks, impacting both the development of secure algorithms and the evaluation of system performance. The development of more realistic adversary models and effective defenses is a key challenge in ensuring the trustworthiness of these increasingly prevalent technologies.

Papers

March 21, 2024

Adversary-Augmented Simulation to evaluate fairness on HyperLedger Fabric
Erwan Mahe, Rouwaida Abdallah, Sara Tucci-Piergiovanni, Pierre-Yves Piriou
Adversarial Attack Procedural Fairness Adversarial Behavior Blockchain Network Hyperledger Fabric Adversary Model

March 10, 2024

Fake or Compromised? Making Sense of Malicious Clients in Federated Learning
Hamid Mozaffari, Sunav Choudhary, Amir Houmansadr
Federated Learning Fake Speech Multiple Sens Malicious Client Adversary Model Aggregation Defense

June 6, 2023

Avoid Adversarial Adaption in Federated Learning by Multi-Metric Investigations
Torsten Krauß, Alexandra Dmitrienko
Adaptive Attack Adaptive Adversary Adversarial Adaptation Adversary Model

April 1, 2023

Coordinated Defense Allocation in Reach-Avoid Scenarios with Efficient Online Optimization
Junwei Liu, Zikai Ouyang, Jiahui Yang, Hua Chen, Haibo Lu, Wei Zhang
Online Convex Optimization Online Optimization Defense Algorithm Reach Avoid Non Faulty Robot Adversary Model

November 28, 2021

Detecting Adversaries, yet Faltering to Noise? Leveraging Conditional Variational AutoEncoders for Adversary Detection in the Presence of Noisy Images
Dvij Kalaria, Aritra Hazra, Partha Pratim Chakrabarti
Deep Neural Network Adversarial Attack Deep Learning Model Industrial Disturbing Noise Noisy Image Conditional Variational Image Perturbation Adversary Model

November 16, 2021

Adversarially Constructed Evaluation Sets Are More Challenging, but May Not Be Fair
Jason Phang, Angelica Chen, William Huang, Samuel R. Bowman
Adversarial Text Evaluation Datasets Evaluation Set Challenging Dataset Adversarial Datasets Adversary Model