Game Redesign in No-regret Game Playing

Authors: Yuzhe Ma, Young Wu, Xiaojin Zhu

IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Simulations on four classic games confirm the effectiveness of our proposed redesign algorithms.
Researcher Affiliation Academia Yuzhe Ma , Young Wu , Xiaojin Zhu University of Wisconsin Madison {yzm234, yw, jerryzhu}@cs.wisc.edu
Pseudocode Yes Algorithm 1 Interior Design; Algorithm 2 Boundary Design; Algorithm 3 Discrete Design
Open Source Code No The paper does not provide any statement about releasing source code or a link to a code repository.
Open Datasets No The paper defines and simulates classic game theory scenarios (e.g., Volunteer's Dilemma, RPS) within the paper itself or by reference to well-known theoretical constructs. It does not use or link to external, publicly available datasets in the traditional sense that require concrete access information.
Dataset Splits No The paper describes running simulations for a number of rounds (T) and analyzing the outcomes, rather than using a train/validation/test split of a dataset.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments, such as GPU/CPU models or cloud resources.
Software Dependencies No The paper mentions using 'EXP3.P [Bubeck and Cesa-Bianchi, 2012]' as the no-regret learner but does not specify a version number for it or for any other key software components or libraries used.
Experiment Setup Yes The margin parameter is ρ = 1. We ran Algorithm 2 for ϵ = 0.1, 0.2, 0.3, 0.4. For each ϵ we simulated game play for T = 104, 105, 106 and 107.