Coordination in Multi-Player Human-Computer Groups

Yaniv Mazliah and Ya’akov (Kobi) Gal

Abstract: There is inconclusive evidence whether practising tasks with computer agents improves people’s performance on these tasks. This talk studies this question empirically using extensive experiments involving human subjects which were trained by computer agents to play a three-player coordination task that is a common test bed in AI to evaluate computational cooperation strategies. Following training, we compared the performance of subjects when playing state-of-the-art agents from the literature. The results revealed that training with computer agents increased people’s performance when compared to the state-of-the-art agent. These results demonstrate the efficacy of using computer agents as tools for improving people’s skills when interacting in strategic settings, saving considerable effort and providing better performance than when training with human counterparts.

Slides: Gal.pptx

Paper: kobi_gal.pdf