Human-Algorithm Collaboration: Achieving Complementarity and Avoiding Unfairness

TL;DR

This paper develops a theoretical framework for human-algorithm collaboration, analyzing conditions for complementarity and fairness, and providing insights to improve collaborative system design.

Contribution

It introduces a formal model for human-algorithm systems, proves when complementarity is possible or impossible, and discusses fairness implications.

Findings

Identifies conditions where complementarity cannot be achieved.

Provides constructive examples of effective human-algorithm collaboration.

Discusses fairness considerations in classifier design.

Abstract

Much of machine learning research focuses on predictive accuracy: given a task, create a machine learning model (or algorithm) that maximizes accuracy. In many settings, however, the final prediction or decision of a system is under the control of a human, who uses an algorithm's output along with their own personal expertise in order to produce a combined prediction. One ultimate goal of such collaborative systems is "complementarity": that is, to produce lower loss (equivalently, greater payoff or utility) than either the human or algorithm alone. However, experimental results have shown that even in carefully-designed systems, complementary performance can be elusive. Our work provides three key contributions. First, we provide a theoretical framework for modeling simple human-algorithm systems and demonstrate that multiple prior analyses can be expressed within it. Next, we use this model to prove conditions where complementarity is impossible, and give constructive examples of where complementarity is achievable. Finally, we discuss the implications of our findings, especially with respect to the fairness of a classifier. In sum, these results deepen our understanding of key factors influencing the combined performance of human-algorithm systems, giving insight into how algorithmic tools can best be designed for collaborative environments.