Debiasing Methods for Fairer Neural Models in Vision and Language Research: A Survey

Abstract

Despite being responsible for state-of-the-art results in several computer vision and natural language processing tasks, neural networks have faced harsh criticism due to some of their current shortcomings. One of them is that neural networks are correlation machines prone to model biases within the data instead of focusing on actual useful causal relationships. This problem is particularly serious in application domains affected by aspects such as race, gender, and age. To prevent models from incurring on unfair decision-making, the AI community has concentrated efforts in correcting algorithmic biases, giving rise to the research area now widely known as fairness in AI. In this survey paper, we provide an in-depth overview of the main debiasing methods for fairness-aware neural networks in the context of vision and language research. We propose a novel taxonomy to better organize the literature on debiasing methods for fairness, and we discuss the current challenges, trends, and important future work directions for the interested researcher and practitioner.

Submitted to ACM Computing Surveys - Special Issue on Trustworthy AI

References in corpus (26)

• Distilling the Knowledge in a Neural Network
• Hierarchical Text-Conditional Image Generation with CLIP Latents
• On the Opportunities and Risks of Foundation Models
• Multi-Task Learning for Dense Prediction Tasks: A Survey
• Improving fairness in machine learning systems: What do industry practitioners need?
• Fairness in Machine Learning: A Survey
• Multi-Task Learning with Deep Neural Networks: A Survey
• Detecting Emergent Intersectional Biases: Contextualized Word Embeddings Contain a Distribution of Human-like Biases
• Predictive Biases in Natural Language Processing Models: A Conceptual Framework and Overview
• Institutionalising Ethics in AI through Broader Impact Requirements
• Societal Biases in Retrieved Contents: Measurement Framework and Adversarial Mitigation for BERT Rankers
• LiFT: A Scalable Framework for Measuring Fairness in ML Applications
• Unpacking the Expressed Consequences of AI Research in Broader Impact Statements
• A Framework for Fairness: A Systematic Review of Existing Fair AI Solutions
• Towards Standardization of Data Licenses: The Montreal Data License
• Measuring Fairness with Biased Rulers: A Survey on Quantifying Biases in Pretrained Language Models
• Rawlsian Fair Adaptation of Deep Learning Classifiers
• Reducing Gender Bias in Word-Level Language Models with a Gender-Equalizing Loss Function
• A Survey on Bias and Fairness in Natural Language Processing
• Conditional Contrastive Learning for Improving Fairness in Self-Supervised Learning
• Rethinking Pareto Frontier for Performance Evaluation of Deep Neural Networks
• MDR Cluster-Debias: A Nonlinear WordEmbedding Debiasing Pipeline
• FairStyle: Debiasing StyleGAN2 with Style Channel Manipulations
• A Source-Criticism Debiasing Method for GloVe Embeddings
• Dataset Bias Mitigation Through Analysis of CNN Training Scores
• [RE] Double-Hard Debias: Tailoring Word Embeddings for Gender Bias Mitigation