Adversarial Traffic

Adversarial traffic research focuses on designing and defending against malicious network traffic crafted to deceive machine learning models used in various applications, such as autonomous driving and network intrusion detection. Current research explores methods for generating adversarial examples, often using gradient-based attacks or probabilistic modeling of traffic patterns, and developing robust defenses, including adversarial training, input perturbation, and ensemble methods. This field is crucial for ensuring the reliability and security of AI systems deployed in safety-critical and security-sensitive environments, impacting the development of robust and trustworthy machine learning models.

Papers

May 16, 2024

Online bipartite matching with imperfect advice
Davin Choo, Themis Gouleakis, Chun Kai Ling, Arnab Bhattacharyya
Online Service Ranking Model Goal Vertex Advice Quality Adversarial Traffic Optimal Offline

October 19, 2023

CAT: Closed-loop Adversarial Training for Safe End-to-End Driving
Linrui Zhang, Zhenghao Peng, Quanyi Li, Bolei Zhou
Autonomous Vehicle Adversarial Scenario End to End Driving Adversarial Traffic

September 4, 2023

Adaptive Resource Allocation for Virtualized Base Stations in O-RAN with Online Learning
Michail Kalntis, George Iosifidis, Fernando A. Kuipers
Online Learning Resource Allocation Online Learning Algorithm Worst Case User Throughput Open Radio Access Network Base Station Optimality Gap Adversarial Traffic

June 6, 2023

Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filters
Xingxing Wei, Shiji Zhao, Bo li
Adversarial Attack Native Robustness Adversarial Example Adversarial Robustness Adversarial Text Max Filter Weight Distribution Adversarial Traffic

April 25, 2023

Zero-shot Transfer Learning of Driving Policy via Socially Adversarial Traffic Flow
Dongkun Zhang, Jintao Xue, Yuxiang Cui, Yunkai Wang, Eryun Liu, Wei Jing, Junbo Chen, Rong Xiong, Yue Wang
Social Navigation Health Policy Recommendation Observable Stochastic Game Dense Traffic Adversarial Traffic

October 8, 2022

FedDef: Defense Against Gradient Leakage in Federated Learning-based Network Intrusion Detection Systems
Jiahui Chen, Yi Zhao, Qi Li, Xuewei Feng, Ke Xu
Adversarial Attack Federated Learning Gradient Leakage Robust Federated Learning Adversarial Traffic

October 5, 2022

Practical Adversarial Attacks on Spatiotemporal Traffic Forecasting Models
Fan Liu, Hao Liu, Wenzhao Jiang
Traffic Forecasting Adversarial Scenario Spatiotemporal Traffic Adversarial Traffic

March 18, 2022

AdIoTack: Quantifying and Refining Resilience of Decision Tree Ensemble Inference Models against Adversarial Volumetric Attacks on IoT Networks
Arman Pashamokhtari, Gustavo Batista, Hassan Habibi Gharakheili
Adversarial Attack Decision Tree Tiny Refinement Elicit Resilience Internet of Thing Network Adversarial Traffic Volumetric Attack

December 22, 2021

Detect & Reject for Transferability of Black-box Adversarial Attacks Against Network Intrusion Detection Systems
Islam Debicha, Thibault Debatty, Jean-Michel Dricot, Wim Mees, Tayeb Kenaza
Adversarial Attack Task Transferability Detection Model Network Intrusion Detection Adversarial Image Black Box Adversarial Attack Risky Prompt Rejection Learning Based Intrusion Detection Adversarial Traffic

Adversarial Traffic

Papers

Online bipartite matching with imperfect advice

CAT: Closed-loop Adversarial Training for Safe End-to-End Driving

Adaptive Resource Allocation for Virtualized Base Stations in O-RAN with Online Learning

Revisiting the Trade-off between Accuracy and Robustness via Weight Distribution of Filters

Zero-shot Transfer Learning of Driving Policy via Socially Adversarial Traffic Flow

FedDef: Defense Against Gradient Leakage in Federated Learning-based Network Intrusion Detection Systems

Practical Adversarial Attacks on Spatiotemporal Traffic Forecasting Models

AdIoTack: Quantifying and Refining Resilience of Decision Tree Ensemble Inference Models against Adversarial Volumetric Attacks on IoT Networks

Detect & Reject for Transferability of Black-box Adversarial Attacks Against Network Intrusion Detection Systems