DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learning

TL;DR

This paper presents DADI, a reinforcement learning framework that dynamically selects information features to optimize fairness and accuracy in predictive models, adaptable to various fairness objectives.

Contribution

The paper introduces a novel adversarial reinforcement learning approach for dynamic feature discovery that balances fairness and accuracy, extending previous adversarial representation learning methods.

Findings

Successfully balances fairness and predictive performance.

Demonstrates effectiveness on real-world datasets.

Provides a flexible framework for fair information discovery.

Abstract

We introduce a framework for dynamic adversarial discovery of information (DADI), motivated by a scenario where information (a feature set) is used by third parties with unknown objectives. We train a reinforcement learning agent to sequentially acquire a subset of the information while balancing accuracy and fairness of predictors downstream. Based on the set of already acquired features, the agent decides dynamically to either collect more information from the set of available features or to stop and predict using the information that is currently available. Building on previous work exploring adversarial representation learning, we attain group fairness (demographic parity) by rewarding the agent with the adversary's loss, computed over the final feature set. Importantly, however, the framework provides a more general starting point for fair or private dynamic information discovery. Finally, we demonstrate empirically, using two real-world datasets, that we can trade-off fairness and predictive performance