在人工智能协助下的人类对齐决策学习
Learning to Decide with AI Assistance under Human-Alignment
AI总结：
本文研究了在高风险领域中，人工智能如何通过预测结果帮助决策者，并探讨了AI预测信心与决策者自身信心的对齐程度对决策学习复杂性的影响。
摘要：
人们普遍认为，当人工智能模型通过预测感兴趣的结果来协助决策者时，它们应传达预测的置信度。然而，实证证据表明，决策者往往难以仅根据传达的置信度来判断何时信任预测。在此背景下，近期的理论和实证工作表明，AI辅助决策的效用与AI置信度和决策者自身置信度之间的对齐程度之间存在正相关性。关键的是，这些发现尚未阐明这种对齐程度如何影响通过重复交互学习做出最佳决策的复杂性。在本文中，我们考虑二元预测和二元决策的典型情况，首先证明该问题等价于具有完全反馈的双臂在线上下文学习问题，并建立了任何学习者可以达到的期望遗憾的下界为$Ω(\sqrt{|H| \cdot |B| \cdot T} )$，其中$H$和$B$分别表示人类和AI置信度的集合。然后我们证明，在AI和人类置信度完全对齐的情况下，学习者可以达到期望遗憾为$O(\sqrt{|H| \cdot T\log T})$，当$\sqrt{|H|} = O(\log T)$且$B$是可数的时，Dvoretzky-Kiefer-Wolfowitz不等式的非平凡推广将遗憾界改进到$O(\sqrt{T\log T})$。这些结果表明，对齐可以减少在人工智能协助下学习决策的复杂性。在两个不同的人类主体研究中，参与者通过AI模型协助解决简单决策任务的实验证明，我们的理论结果在完全对齐被违反时仍然稳健。
英文摘要：
It is widely agreed that when AI models assist decision-makers in high-stakes domains by predicting an outcome of interest, they should communicate the confidence of their predictions. However, empirical evidence suggests that decision-makers often struggle to determine when to trust a prediction based solely on this communicated confidence. In this context, recent theoretical and empirical work suggests a positive correlation between the utility of AI-assisted decision-making and the degree of alignment between the AI confidence and the decision-makers' confidence in their own predictions. Crucially, these findings do not yet elucidate the extent to which this alignment influences the complexity of learning to make optimal decisions through repeated interactions. In this paper, we address this question in the canonical case of binary predictions and binary decisions. We first show that this problem is equivalent to a two-armed online contextual learning problem with full feedback, and establish a lower bound of $Ω(\sqrt{|H| \cdot |B| \cdot T} )$ on the expected regret any learner can attain, where $H$ and $B$ denote the sets of human and AI confidence values. We then demonstrate that, under perfect alignment between AI and human confidence, a learner can attain an expected regret of $O(\sqrt{|H| \cdot T\log T})$ and, when $\sqrt{|H|} = O(\log T)$ and $B$ is countable, a non-trivial generalization of the Dvoretzky-Kiefer-Wolfowitz inequality improves the regret bound to $O(\sqrt{T\log T})$. Taken together, these results reveal that alignment can reduce the complexity of learning to make decisions with AI assistance. Experiments on real data from two different human-subject studies where participants solve simple decision-making tasks assisted by AI models show that our theoretical results are robust to violations of perfect alignment.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
