Skill Rating for Generative Models
arXiv: Machine Learning, Volume abs/1808.04888, 2018.
EI
Abstract:
We explore a new way to evaluate generative models using insights from evaluation of competitive games between human players. We show experimentally that tournaments between generators and discriminators provide an effective way to evaluate generative models. We introduce two methods for summarizing tournament outcomes: tournament win rat...More
Code:
Data:
Tags
Comments