研究人类如何评价AI队友，发现行为多样性比得分更重要。
In Pursuit of Predictive Models of Human Preferences Toward AI Teammates
- 用汉诺比游戏测试AI协作表现，结合客观指标与人类偏好。
- 人类更看重AI行为多样性、战略主导性和协同能力，而非最终得分。
- 结果挑战传统强化学习假设，适合训练人机协作AI的开发者参考。
我们探索影响人类对AI队友主观评价的可测量特征。实验采用合作类卡牌游戏汉诺比（Hanabi）作为常见基准。首先，基于任务表现、信息论和博弈论评估AI代理的客观指标，这些指标无需人类交互即可计算。随后，在大规模（N=241）人机协作实验中评估人类对AI队友的主观偏好。最后，将仅依赖AI的客观指标与人类偏好进行相关性分析。结果反驳了强化学习领域既有的常见假设，揭示出新的关联：人类偏好与团队最终得分的相关性较低，反而与AI行为多样性、战略主导性以及与其他AI协作的能力密切相关。未来这些相关性或可指导训练人机协作型AI的奖励函数设计。
原文摘要 · Abstract (English)
We seek measurable properties of AI agents that make them better or worse teammates from the subjective perspective of human collaborators. Our experiments use the cooperative card game Hanabi -- a common benchmark for AI-teaming research. We first evaluate AI agents on a set of objective metrics based on task performance, information theory, and game theory, which are measurable without human interaction. Next, we evaluate subjective human preferences toward AI teammates in a large-scale (N=241) human-AI teaming experiment. Finally, we correlate the AI-only objective metrics with the human subjective preferences. Our results refute common assumptions from prior literature on reinforcement learning, revealing new correlations between AI behaviors and human preferences. We find that the final game score a human-AI team achieves is less predictive of human preferences than esoteric measures of AI action diversity, strategic dominance, and ability to team with other AI. In the future, these correlations may help shape reward functions for training human-collaborative AI.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。