Statistical discrimination in learning agents

Abstract

Undesired bias afflicts both human and algorithmic decision making, and may be especially prevalent when information processing trade-offs incentivize the use of heuristics. One primary example is \textit{statistical discrimination} -- selecting social partners based not on their underlying attributes, but on readily perceptible characteristics that covary with their suitability for the task at hand. We present a theoretical model to examine how information processing influences statistical discrimination and test its predictions using multi-agent reinforcement learning with various agent architectures in a partner choice-based social dilemma. As predicted, statistical discrimination emerges in agent policies as a function of both the bias in the training population and of agent architecture. All agents showed substantial statistical discrimination, defaulting to using the readily available correlates instead of the outcome relevant features. We show that less discrimination emerges with agents that use recurrent neural networks, and when their training environment has less bias. However, all agent algorithms we tried still exhibited substantial bias after learning in biased training populations.

arXiv comment29 pages, 10 figures

References in corpus (7)

• Semantics derived automatically from language corpora contain human-like biases
• On Formalizing Fairness in Prediction with Machine Learning
• Discrimination, artificial intelligence, and algorithmic decision-making
• Quantifying the Carbon Emissions of Machine Learning
• Maintaining cooperation in complex social dilemmas using deep reinforcement learning
• V-MPO: On-Policy Maximum a Posteriori Policy Optimization for Discrete and Continuous Control
• Model-free conventions in multi-agent reinforcement learning with heterogeneous preferences