Bias, Fairness, and Accountability with AI and ML Algorithms
arXiv:2105.06558
Abstract
The advent of AI and ML algorithms has led to opportunities as well as challenges. In this paper, we provide an overview of bias and fairness issues that arise with the use of ML algorithms. We describe the types and sources of data bias, and discuss the nature of algorithmic unfairness. This is followed by a review of fairness metrics in the literature, discussion of their limitations, and a description of de-biasing (or mitigation) techniques in the model life cycle.
arXiv comment18 pages, 5 figures
References in corpus (17)
- Axiomatic Attribution for Deep Networks
- Learning Important Features Through Propagating Activation Differences
- Equality of Opportunity in Supervised Learning
- Predictive learning via rule ensembles
- Consistent Individualized Feature Attribution for Tree Ensembles
- The Measure and Mismeasure of Fairness
- Fairness Constraints: Mechanisms for Fair Classification
- On Fairness and Calibration
- Learning Non-Discriminatory Predictors
- A Reductions Approach to Fair Classification
- Explainable Neural Networks based on Additive Index Models
- A Debiased MDI Feature Importance Measure for Random Forests
- Aequitas: A Bias and Fairness Audit Toolkit
- Locally Interpretable Models and Effects based on Supervised Partitioning (LIME-SUP)
- On conditional parity as a notion of non-discrimination in machine learning
- Enhancing Explainability of Neural Networks through Architecture Constraints
- Model Interpretation: A Unified Derivative-based Framework for Nonparametric Regression and Supervised Machine Learning