DeepAI AI Chat
Log In Sign Up

Statistical discrimination in learning agents

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 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.


page 2

page 7


A Multi-Agent Model for Polarization under Confirmation Bias in Social Networks

We describe a model for polarization in multi-agent systems based on Est...

A Simulation Based Dynamic Evaluation Framework for System-wide Algorithmic Fairness

We propose the use of Agent Based Models (ABMs) inside a reinforcement l...

On Statistical Discrimination as a Failure of Social Learning: A Multi-Armed Bandit Approach

We analyze statistical discrimination using a multi-armed bandit model w...

The Bias-Expressivity Trade-off

Learning algorithms need bias to generalize and perform better than rand...

Energy-based Surprise Minimization for Multi-Agent Value Factorization

Multi-Agent Reinforcement Learning (MARL) has demonstrated significant s...

Optimized Data Pre-Processing for Discrimination Prevention

Non-discrimination is a recognized objective in algorithmic decision mak...

Shepherding Heterogeneous Flock with Model-Based Discrimination

The problem of guiding a flock of agents to a destination by the repulsi...