AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias

Abstract

Fairness is an increasingly important concern as machine learning models are used to support decision making in high-stakes applications such as mortgage lending, hiring, and prison sentencing. This paper introduces a new open source Python toolkit for algorithmic fairness, AI Fairness 360 (AIF360), released under an Apache v2.0 license {https://github.com/ibm/aif360). The main objectives of this toolkit are to help facilitate the transition of fairness research algorithms to use in an industrial setting and to provide a common framework for fairness researchers to share and evaluate algorithms. The package includes a comprehensive set of fairness metrics for datasets and models, explanations for these metrics, and algorithms to mitigate bias in datasets and models. It also includes an interactive Web experience (https://aif360.mybluemix.net) that provides a gentle introduction to the concepts and capabilities for line-of-business users, as well as extensive documentation, usage guidance, and industry-specific tutorials to enable data scientists and practitioners to incorporate the most appropriate tool for their problem into their work products. The architecture of the package has been engineered to conform to a standard paradigm used in data science, thereby further improving usability for practitioners. Such architectural design and abstractions enable researchers and developers to extend the toolkit with their new algorithms and improvements, and to use it for performance benchmarking. A built-in testing infrastructure maintains code quality.

20 pages

References in corpus (6)

• Equality of Opportunity in Supervised Learning
• Counterfactual Fairness
• Actionable Recourse in Linear Classification
• Fairness Testing: Testing Software for Discrimination
• On Fairness and Calibration
• On the (im)possibility of fairness

Cited by in corpus (88)

• The What-If Tool: Interactive Probing of Machine Learning Models
• Fairness in Machine Learning: A Survey
• An Empirical Characterization of Fair Machine Learning For Clinical Risk Prediction
• Fairway: A Way to Build Fair ML Software
• Do the Machine Learning Models on a Crowd Sourced Platform Exhibit Bias? An Empirical Study on Model Fairness
• Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
• Recommendations on test datasets for evaluating AI solutions in pathology
• Rule Extraction in Unsupervised Anomaly Detection for Model Explainability: Application to OneClass SVM
• Principles to Practices for Responsible AI: Closing the Gap
• Aequitas: A Bias and Fairness Audit Toolkit
• FlipTest: Fairness Testing via Optimal Transport
• In AI We Trust? Factors That Influence Trustworthiness of AI-infused Decision-Making Processes
• LiFT: A Scalable Framework for Measuring Fairness in ML Applications
• On the Moral Justification of Statistical Parity
• Mitigating Bias in Federated Learning
• Fairness for Robust Log Loss Classification
• Socially Responsible AI Algorithms: Issues, Purposes, and Challenges
• dalex: Responsible Machine Learning with Interactive Explainability and Fairness in Python
• Fairness via Representation Neutralization
• Can Explainable AI Explain Unfairness? A Framework for Evaluating Explainable AI
• End-To-End Bias Mitigation: Removing Gender Bias in Deep Learning
• FairPrep: Promoting Data to a First-Class Citizen in Studies on Fairness-Enhancing Interventions
• Question-Driven Design Process for Explainable AI User Experiences
• Explainability for fair machine learning
• FACT: A Diagnostic for Group Fairness Trade-offs
• Data Quality Toolkit: Automatic assessment of data quality and remediation for machine learning datasets
• "What We Can't Measure, We Can't Understand": Challenges to Demographic Data Procurement in the Pursuit of Fairness
• Ensuring Fairness Beyond the Training Data
• Pretrained AI Models: Performativity, Mobility, and Change
• Estimating and Improving Fairness with Adversarial Learning
• A fuzzy-rough uncertainty measure to discover bias encoded explicitly or implicitly in features of structured pattern classification datasets
• Software Engineering for Fairness: A Case Study with Hyperparameter Optimization
• FAT Forensics: A Python Toolbox for Algorithmic Fairness, Accountability and Transparency
• Foundations of Explainable Knowledge-Enabled Systems
• BeFair: Addressing Fairness in the Banking Sector
• Removing biased data to improve fairness and accuracy
• Fairness in Algorithmic Profiling: A German Case Study
• Tight Mutual Information Estimation With Contrastive Fenchel-Legendre Optimization
• Fairness On The Ground: Applying Algorithmic Fairness Approaches to Production Systems
• Balanced Ranking with Diversity Constraints
• Biases in Data Science Lifecycle
• Fairness-Aware Neural Réyni Minimization for Continuous Features
• Approaching Machine Learning Fairness through Adversarial Network
• Fair Classification with Noisy Protected Attributes: A Framework with Provable Guarantees
• Mitigating Dataset Harms Requires Stewardship: Lessons from 1000 Papers
• Detection and Mitigation of Bias in Ted Talk Ratings
• Fairkit, Fairkit, on the Wall, Who's the Fairest of Them All? Supporting Data Scientists in Training Fair Models
• Modelling the long-term fairness dynamics of data-driven targeted help on job seekers
• Understanding racial bias in health using the Medical Expenditure Panel Survey data
• Personalized Chatbot Trustworthiness Ratings
• Link Prediction using Graph Neural Networks for Master Data Management
• Developing Future Human-Centered Smart Cities: Critical Analysis of Smart City Security, Interpretability, and Ethical Challenges
• On the Privacy Risks of Algorithmic Fairness
• A Reliability-aware Distributed Framework to Schedule Residential Charging of Electric Vehicles
• Data Cleaning for Accurate, Fair, and Robust Models: A Big Data - AI Integration Approach
• Designing Evaluations of Machine Learning Models for Subjective Inference: The Case of Sentence Toxicity
• Fairness in Forecasting of Observations of Linear Dynamical Systems
• Fairness and Missing Values
• Evidence-based explanation to promote fairness in AI systems
• FairMask: Better Fairness via Model-based Rebalancing of Protected Attributes
• Metrics and methods for a systematic comparison of fairness-aware machine learning algorithms
• Addressing Cognitive Biases in Augmented Business Decision Systems
• Bias Mitigation Post-processing for Individual and Group Fairness
• Towards Fairness in Classifying Medical Conversations into SOAP Sections
• Uncovering the Bias in Facial Expressions
• Linear Classifiers that Encourage Constructive Adaptation
• Know Your Model (KYM): Increasing Trust in AI and Machine Learning
• Deep AUC Maximization for Medical Image Classification: Challenges and Opportunities
• Transitioning from Real to Synthetic data: Quantifying the bias in model
• Statistical inference for individual fairness
• Quality Management of Machine Learning Systems
• Towards a Measure of Individual Fairness for Deep Learning
• Fairness in Deep Learning: A Computational Perspective
• Analyzing Bias in Sensitive Personal Information Used to Train Financial Models
• Profit equitably: An investigation of market maker's impact on equitable outcomes
• One-vs.-One Mitigation of Intersectional Bias: A General Method to Extend Fairness-Aware Binary Classification
• Faking Fairness via Stealthily Biased Sampling
• Making ML models fairer through explanations: the case of LimeOut
• Trusting Fair Data: Leveraging Quality in Fairness-Driven Data Removal Techniques
• DataExposer: Exposing Disconnect between Data and Systems
• Towards Integrating Fairness Transparently in Industrial Applications
• Unfairness towards subjective opinions in Machine Learning
• Selective Regression Under Fairness Criteria
• Individually Fair Gradient Boosting
• Do's and Don'ts for Human and Digital Worker Integration
• Towards Data-Driven Affirmative Action Policies under Uncertainty
• Legally grounded fairness objectives
• Slice Tuner: A Selective Data Acquisition Framework for Accurate and Fair Machine Learning Models