Learning in Multi-Memory Games Triggers Complex Dynamics Diverging from Nash Equilibrium

Authors: Yuma Fujimoto, Kaito Ariu, Kenshi Abe

IJCAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Then, we prove their dynamics are identical. Furthermore, theoretically and experimentally, we clarify that the learning dynamics diverge from the Nash equilibrium in multi-memory zero-sum games and reach heteroclinic cycles (sojourn longer around the boundary of the strategy space), providing a fundamental advance in learning in games.
Researcher Affiliation Collaboration Yuma Fujimoto1,2,3 , Kaito Ariu3,4 and Kenshi Abe3 1Research Center for Integrative Evolutionary Science, SOKENDAI. 2Universal Biology Institute (UBI), the University of Tokyo. 3AI Lab, Cyber Agent, Inc. 4KTH Royal Institute of Technology.
Pseudocode Yes Algorithm 1 Discretized MMRD Input: η 1: for t = 0, 1, 2, do ... Algorithm 2 Discretized MMGA Input: η, γ 1: for t = 0, 1, 2, do ...
Open Source Code Yes The codes that we used are available at https://github.com/ Cyber Agent AILab/with-memory games
Open Datasets No The paper describes standard game theory setups (e.g., matching-pennies game, rock-paper-scissors game) which are defined within the paper, but it does not use or provide access information for a separate, publicly available dataset.
Dataset Splits No The paper describes simulations of game theory dynamics and does not refer to traditional training/validation/test dataset splits.
Hardware Specification No The paper does not specify any hardware used for running the experiments or simulations.
Software Dependencies No The paper does not list any specific software dependencies with version numbers.
Experiment Setup Yes The step-size is 10^-2 in common. ... We use Algorithm 2 with η = 10^-3 and γ = 10^-6. ... where we use Algorithm 2 with η = 10^-2 and γ = 10^-6.