NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:7259
Title:Self-supervised GAN: Analysis and Improvement with Multi-class Minimax Game


		
The paper addresses a problem in self supervised GAN, where the classes strictly have disjoint support. This is mitigated by introducing a new class for generated samples. Experiments are convincing.