Reproducibility Challenge NeurIPS 2019 Report on Competitive Gradient Descent

Abstract

This is a report for reproducibility challenge of NeurlIPS’19 on the paper Competitive gradient descent (Schäfer et al., 2019). The paper introduces a novel algorithm for the numerical computation of Nash equilibria of competitive two-player games. It avoids oscillatory and divergent behaviors seen in alternating gradient descent. The purpose of this report is to critically examine the reproducibility of the work by (Schäfer et al., 2019), within the framework of the NeurIPS 2019 Reproducibility Challenge. The experiments replicated in this report confirms the results of the original study. Moreover, this project offers a Python(Pytorch based) implementation of the proposed CGD algorithm which can be found at the GitHub public git repository.

Download arXiv version of paper from here

Github code

Recommended citation: kishan, G. (2020). “Reproducibility Challenge NeurIPS 2019 Report on “Competitive Gradient Descent.” Reproducibility Challenge NeurIPS 2019.