[3A1-GS-10] AI application:

[3A1-GS-10-04] Analysis of the El Farol Bar Problem by LLM Agents

〇Ryosuke Takata¹, Atsushi Masumori¹, Takashi Ikegami¹ (1. University of Tokyo)

Keywords:LLM, El Farol Bar Problem

We simulated the autonomous solution of the spatially extended El Farol Bar problem by a group of Large Language Model (LLM) based agents. As a result, the LLM agents generated a spontaneous motivation to go to the bar and changed their decision making by becoming a collective. We also observed that the LLM agents did not solve the problem completely, but rather behaved more like humans. These findings suggest that a new model of group decision making, which could not be handled in the previous game-theoretic problem setting, can be realized by LLM agents.