Biased Training

Biased training in machine learning refers to the phenomenon where models learn spurious correlations from skewed or unrepresentative training data, leading to unfair or inaccurate predictions for certain subgroups. Current research focuses on mitigating this bias through various techniques, including data augmentation with synthetic samples, developing bias-aware training algorithms (e.g., those adjusting sample weights or employing contrastive learning), and post-hoc methods that modify model predictions or architectures to improve fairness. Addressing biased training is crucial for ensuring the reliability and ethical deployment of machine learning models across diverse applications, particularly in sensitive domains like healthcare and finance.

Papers

November 18, 2022

Explanation on Pretraining Bias of Finetuned Vision Transformer
Bumjin Park, Jaesik Choi
Vision Transformer Line by Line Explanation Attention Map Attention Pattern Biased Training Vector Attention Attribution Mask

November 4, 2022

Biased Self-supervised learning for ASR
Florian L. Kreyssig, Yangyang Shi, Jinxi Guo, Leda Sari, Abdelrahman Mohamed, Philip C. Woodland
Self Supervised Learning Automatic Speech Recognition Self Supervised Biased Training

October 18, 2022

Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face Recognition
Samuel Dooley, Rhea Sanjay Sukthanker, John P. Dickerson, Colin White, Frank Hutter, Micah Goldblum
Face Recognition Bias Mitigation Model Bias Efficient Architecture Biased Training

October 11, 2022

Training Debiased Subnetworks with Contrastive Weight Pruning
Geon Yeong Park, Sangmin Lee, Sang Wan Lee, Jong Chul Ye
Training Data Debiasing Method Biased Training Context Pruning

October 10, 2022

Extracting or Guessing? Improving Faithfulness of Event Temporal Relation Extraction
Haoyu Wang, Hongming Zhang, Yuqian Deng, Jacob R. Gardner, Dan Roth, Muhao Chen
Temporal Relation Event Representation Label Bias Biased Training Extraction Model

July 7, 2022

An Exploration of How Training Set Composition Bias in Machine Learning Affects Identifying Rare Objects
Sean E. Lake, Chao-Wei Tsai
Machine Learning Absolute Stance Bias Classification Model Biased Training Rare Class Rare Object

June 7, 2022

Reconsidering Learning Objectives in Unbiased Recommendation with Unobserved Confounders
Teng Xiao, Zhengyu Chen, Suhang Wang
Unobserved Confounders Unbiased Learning Biased Training Biased Feedback Debiased Recommendation

May 17, 2022

Unbiased Math Word Problems Benchmark for Mitigating Solving Bias
Zhicheng Yang, Jinghui Qin, Jiaqi Chen, Xiaodan Liang
Math Word Problem Biased Training Electromagnetic Simulation Math Benchmark

February 15, 2022

Debiased Self-Training for Semi-Supervised Learning
Baixu Chen, Junguang Jiang, Ximei Wang, Pengfei Wan, Jianmin Wang, Mingsheng Long
Semi Supervised Learning Pseudo Labeling Biased Training

February 11, 2022

CMW-Net: Learning a Class-Aware Sample Weighting Mapping for Robust Deep Learning
Jun Shu, Xiang Yuan, Deyu Meng, Zongben Xu
Class Imbalance Robust Deep Biased Training Weighting Function

December 30, 2021

Delving into Sample Loss Curve to Embrace Noisy and Imbalanced Data
Shenwang Jiang, Jianan Li, Ying Wang, Bo Huang, Zhang Zhang, Tingfa Xu
Imbalanced Data Biased Training Sampling Bias Sample Reweighting Sample Loss

December 29, 2021

EiFFFeL: Enforcing Fairness in Forests by Flipping Leaves
Seyum Assefa Abebe, Claudio Lucchese, Salvatore Orlando
Procedural Fairness Decision Tree Group Fairness Forest Environment Biased Training Achieving Fairness