The Fair Game: Auditing & Debiasing AI Algorithms Over Time

Abstract

An emerging field of AI, namely Fair Machine Learning (ML), aims to quantify different types of bias (also known as unfairness) exhibited in the predictions of ML algorithms, and to design new algorithms to mitigate them. Often, the definitions of bias used in the literature are observational, i.e. they use the input and output of a pre-trained algorithm to quantify a bias under concern. In reality,these definitions are often conflicting in nature and can only be deployed if either the ground truth is known or only in retrospect after deploying the algorithm. Thus,there is a gap between what we want Fair ML to achieve and what it does in a dynamic social environment. Hence, we propose an alternative dynamic mechanism,"Fair Game",to assure fairness in the predictions of an ML algorithm and to adapt its predictions as the society interacts with the algorithm over time. "Fair Game" puts together an Auditor and a Debiasing algorithm in a loop around an ML algorithm. The "Fair Game" puts these two components in a loop by leveraging Reinforcement Learning (RL). RL algorithms interact with an environment to take decisions, which yields new observations (also known as data/feedback) from the environment and in turn, adapts future decisions. RL is already used in algorithms with pre-fixed long-term fairness goals. "Fair Game" provides a unique framework where the fairness goals can be adapted over time by only modifying the auditor and the different biases it quantifies. Thus,"Fair Game" aims to simulate the evolution of ethical and legal frameworks in the society by creating an auditor which sends feedback to a debiasing algorithm deployed around an ML system. This allows us to develop a flexible and adaptive-over-time framework to build Fair ML systems pre- and post-deployment.

References in corpus (33)

• Training language models to follow instructions with human feedback
• Equality of Opportunity in Supervised Learning
• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
• Demystifying the Draft EU Artificial Intelligence Act
• Differentially Private Empirical Risk Minimization
• Fairness in Machine Learning: A Survey
• Fairness Testing: Testing Software for Discrimination
• Fine-Tuning Language Models from Human Preferences
• Cultural Bias and Cultural Alignment of Large Language Models
• Identifying and Correcting Label Bias in Machine Learning
• Aligning Large Language Models with Human: A Survey
• Null Compliance: NYC Local Law 144 and the Challenges of Algorithm Accountability
• BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset
• The implicit fairness criterion of unconstrained learning
• Safe RLHF: Safe Reinforcement Learning from Human Feedback
• Fair Regression with Wasserstein Barycenters
• Capuchin: Causal Database Repair for Algorithmic Fairness
• PrivFairFL: Privacy-Preserving Group Fairness in Federated Learning
• How Biased are Your Features?: Computing Fairness Influence Functions with Global Sensitivity Analysis
• Survey on Fair Reinforcement Learning: Theory and Practice
• Sample-Efficient Reinforcement Learning of Partially Observable Markov Games
• On the convergence of policy gradient methods to Nash equilibria in general stochastic games
• SENTINEL: Taming Uncertainty with Ensemble-based Distributional Reinforcement Learning
• Differential Privacy has Bounded Impact on Fairness in Classification
• Statistical Inference for Fairness Auditing
• Near-Optimal Learning of Extensive-Form Games with Imperfect Information
• Algorithmic audits of algorithms, and the law
• SAAC: Safe Reinforcement Learning as an Adversarial Game of Actor-Critics
• On the Complexity of Differentially Private Best-Arm Identification with Fixed Confidence
• Under manipulations, are some AI models harder to audit?
• Bandits Corrupted by Nature: Lower Bounds on Regret and Robust Optimistic Algorithm
• Adapting to game trees in zero-sum imperfect information games
• Risk-Sensitive Bayesian Games for Multi-Agent Reinforcement Learning under Policy Uncertainty