A Q-learning Approach for Adherence-Aware Recommendations

TL;DR

This paper introduces an adherence-aware Q-learning algorithm that learns human adherence levels to AI recommendations and optimizes decision policies in safety-critical scenarios, ensuring convergence to optimal solutions.

Contribution

The paper presents a novel Q-learning method that dynamically learns adherence levels and adapts recommendations accordingly, addressing decision-making in high-stakes environments.

Findings

Algorithm converges to the optimal value

Effective in various simulated scenarios

Improves decision quality by accounting for adherence

Abstract

In many real-world scenarios involving high-stakes and safety implications, a human decision-maker (HDM) may receive recommendations from an artificial intelligence while holding the ultimate responsibility of making decisions. In this letter, we develop an "adherence-aware Q-learning" algorithm to address this problem. The algorithm learns the "adherence level" that captures the frequency with which an HDM follows the recommended actions and derives the best recommendation policy in real time. We prove the convergence of the proposed Q-learning algorithm to the optimal value and evaluate its performance across various scenarios.