Smooth Fictitious Play in Stochastic Games with Perturbed Payoffs and Unknown Transitions
Authors: Lucas Baudin, Rida Laraki
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | We prove the convergence of our family of procedures to stationary regularized Nash equilibria in zero-sum and identical-interest discounted stochastic games. The proof uses the continuous smooth best-response dynamics counterparts, and stochastic approximation methods. |
| Researcher Affiliation | Academia | Lucas Baudin Université Paris-Dauphine PSL lucas.baudin@dauphine.eu Rida Laraki CNRS, Université Paris-Dauphine PSL University of Liverpool |
| Pseudocode | No | The paper describes mathematical systems and update rules (e.g., equations for ui and xs) but does not present them in a formal pseudocode block or algorithm format. |
| Open Source Code | No | The paper does not contain any explicit statement about releasing source code for the described methodology, nor does it provide links to a code repository. The checklist indicates 'N/A' for questions regarding code availability. |
| Open Datasets | No | The paper primarily presents theoretical results and proofs. While Appendix C mentions an example with empirical results, the paper does not describe the use of any specific public dataset, nor does it provide any links or formal citations to a dataset for training. |
| Dataset Splits | No | The paper focuses on theoretical convergence proofs and does not describe any training, validation, or test dataset splits. The checklist explicitly states 'N/A' for training details including data splits. |
| Hardware Specification | No | The paper is theoretical and does not provide any specific hardware specifications (e.g., GPU/CPU models, memory) used for running experiments. The checklist states 'N/A' for details on compute resources. |
| Software Dependencies | No | The paper is theoretical and does not list any specific software dependencies with version numbers that would be required to replicate experiments. The checklist states 'N/A' for training details. |
| Experiment Setup | No | The paper presents theoretical models and proofs rather than an empirical experimental setup. It does not provide specific hyperparameter values, training configurations, or system-level settings for experiments. The checklist states 'N/A' for training details. |