Fairway: A Way to Build Fair ML Software

Abstract

Machine learning software is increasingly being used to make decisions that affect people's lives. But sometimes, the core part of this software (the learned model), behaves in a biased manner that gives undue advantages to a specific group of people (where those groups are determined by sex, race, etc.). This "algorithmic discrimination" in the AI software systems has become a matter of serious concern in the machine learning and software engineering community. There have been works done to find "algorithmic bias" or "ethical bias" in the software system. Once the bias is detected in the AI software system, the mitigation of bias is extremely important. In this work, we a)explain how ground-truth bias in training data affects machine learning model fairness and how to find that bias in AI software,b)propose a methodFairwaywhich combines pre-processing and in-processing approach to remove ethical bias from training data and trained model. Our results show that we can find bias and mitigate bias in a learned model, without much damaging the predictive performance of that model. We propose that (1) test-ing for bias and (2) bias mitigation should be a routine part of the machine learning software development life cycle. Fairway offers much support for these two purposes.

arXiv commentESEC/FSE'20: The 28th ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering

References in corpus (6)

• A Unified Approach to Interpreting Model Predictions
• Semantics derived automatically from language corpora contain human-like biases
• AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias
• Using Bad Learners to find Good Configurations
• FairTest: Discovering Unwarranted Associations in Data-Driven Applications
• Software Engineering for Fairness: A Case Study with Hyperparameter Optimization

Cited by in corpus (14)

• Bias in Machine Learning Software: Why? How? What to do?
• Fairness-aware Configuration of Machine Learning Libraries
• A survey on datasets for fairness-aware machine learning
• Latent Imitator: Generating Natural Individual Discriminatory Instances for Black-Box Fairness Testing
• Making Fair ML Software using Trustworthy Explanation
• BiasRV: Uncovering Biased Sentiment Predictions at Runtime
• To Be Forgotten or To Be Fair: Unveiling Fairness Implications of Machine Unlearning Methods
• FairBalance: How to Achieve Equalized Odds With Data Pre-processing
• Fix Fairness, Don't Ruin Accuracy: Performance Aware Fairness Repair using AutoML
• Towards an Enhanced Understanding of Bias in Pre-trained Neural Language Models: A Survey with Special Emphasis on Affective Bias
• Fair-SSL: Building fair ML Software with less data
• Unmasking Gender Bias in Recommendation Systems and Enhancing Category-Aware Fairness
• FairGridSearch: A Framework to Compare Fairness-Enhancing Models
• Ideology in Open Source Development