Adversarial Bias

Adversarial bias in machine learning refers to the unintended biases learned by models, often reflecting and amplifying societal prejudices present in training data. Current research focuses on developing methods to detect and mitigate these biases, exploring techniques like adversarial training with auxiliary models to identify bias features without needing explicit bias labels, and investigating the interplay between model architecture, data structure, and robustness to adversarial attacks. Understanding and addressing adversarial bias is crucial for ensuring fairness and reliability in AI systems across diverse applications, from medical diagnosis to environmental monitoring, where biased predictions can have significant real-world consequences.

Papers

August 10, 2024

MABR: A Multilayer Adversarial Bias Removal Approach Without Prior Bias Knowledge
Maxwell J. Yin, Boyu Wang, Charles Ling
Gender Bias Bias Mitigation Social Bias Prior Knowledge Biased Feature Adversarial Bias

May 24, 2024

Can Implicit Bias Imply Adversarial Robustness?
Hancheng Min, René Vidal
Adversarial Attack Adversarial Robustness Implicit Bias Shallow Network Polynomial Activation Adversarial Bias

April 1, 2024

The Double-Edged Sword of Input Perturbations to Robust Accurate Fairness
Xuran Li, Peng Wu, Yanting Chen, Xingjun Ma, Zhen Zhang, Kaixiang Dong
Adversarial Perturbation Input Perturbation Adversarial Instance Robust Fairness Adversarial Bias

February 8, 2024

Hidden in Plain Sight: Undetectable Adversarial Bias Attacks on Vulnerable Patient Populations
Pranav Kulkarni, Andrew Chan, Nithya Navarathna, Skylar Chan, Paul H. Yi, Vishwa S. Parekh
Hidden Knowledge Plain Sight Prediction Bias Vulnerable Population Bias Neuron Adversarial Bias

December 20, 2022

Debiasing Stance Detection Models with Counterfactual Reasoning and Adversarial Bias Learning
Jianhua Yuan, Yanyan Zhao, Bing Qin
Stance Detection Counterfactual Reasoning Dataset Bias Counterfactual Inference Adversarial Bias

September 27, 2022

Inducing Data Amplification Using Auxiliary Datasets in Adversarial Training
Saehyung Lee, Hyungyu Lee
Adversarial Training Adversarial Robustness Multi Domain Learning Auxiliary Data Adversarial Bias Sample Amplification

December 3, 2021

Adversarial Attacks against a Satellite-borne Multispectral Cloud Detector
Andrew Du, Yee Wei Law, Michele Sasdelli, Bo Chen, Ken Clarke, Michael Brown, Tat-Jun Chin
Adversarial Attack Cloud Detection Adversarial Bias