AutoGAN: Neural Architecture Search for Generative Adversarial Networks

ICCV 2019 Xinyu GongShiyu ChangYifan JiangZhangyang Wang

Neural architecture search (NAS) has witnessed prevailing success in image classification and (very recently) segmentation tasks. In this paper, we present the first preliminary study on introducing the NAS algorithm to generative adversarial networks (GANs), dubbed AutoGAN... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Image Generation CIFAR-10 AutoGAN Inception score 8.55 # 3
Image Generation CIFAR-10 AutoGAN FID 12.42 # 2
Neural Architecture Search CIFAR-10 image generation AutoGAN FID 12.42 # 1
Neural Architecture Search CIFAR-10 image generation AutoGAN Inception score 8.55 # 1