Converging to Team-Maxmin Equilibria in Zero-Sum Multiplayer Games

Authors: Youzhi Zhang, Bo An

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, extensive experimental results show that CISGT is orders of magnitude faster than ISGT and the state-of-the-art algorithm to compute TMEs in large games.
Researcher Affiliation Academia 1School of Computer Science and Engineering, Nanyang Technological University, Singapore.
Pseudocode Yes Algorithm 1 ISG for a TME (ISGT) and Algorithm 2 The CTME Based ISGT (CISGT)
Open Source Code No The paper does not provide an explicit statement or link for the open-source code of the described methodology.
Open Datasets Yes We conduct experiments on the classic network security games (Washburn & Wood, 1995; Jain et al., 2011; Iwashita et al., 2016)... All networks are generated by the grid model with random edges (Peng et al., 2014)...
Dataset Splits No The paper mentions generating random instances for experiments but does not explicitly detail training, validation, or test dataset splits (e.g., percentages, sample counts, or predefined splits).
Hardware Specification Yes All experiments are run on a machine with 6-core 3.6GHz CPU and 32GB memory.
Software Dependencies Yes We use CPLEX solver (version 12.9) for solving all linear programs.
Experiment Setup Yes We conduct experiments on the classic network security games... (i.e., ϵ-TME with ϵ = 0.05)... By default, n = 3, and results are all averaged over 30 instances that are randomly generated.