Investigating Under and Overfitting in Wasserstein Generative Adversarial Networks

Abstract

We investigate under and overfitting in Generative Adversarial Networks (GANs), using discriminators unseen by the generator to measure generalization. We find that the model capacity of the discriminator has a significant effect on the generator's model quality, and that the generator's poor performance coincides with the discriminator underfitting. Contrary to our expectations, we find that generators with large model capacities relative to the discriminator do not show evidence of overfitting on CIFAR10, CIFAR100, and CelebA.

References in corpus (9)

• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• NIPS 2016 Tutorial: Generative Adversarial Networks
• A Note on the Inception Score
• Generalization and Equilibrium in Generative Adversarial Nets (GANs)
• The Numerics of GANs
• Do GANs actually learn the distribution? An empirical study
• Approximability of Discriminators Implies Diversity in GANs
• Finding Mixed Nash Equilibria of Generative Adversarial Networks

Cited by in corpus (5)

• On Filter Generalization for Music Bandwidth Extension Using Deep Neural Networks
• How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models
• Local and non-local dependency learning and emergence of rule-like representations in speech data by Deep Convolutional Generative Adversarial Networks
• Generative Minimization Networks: Training GANs Without Competition
• Identity-Based Patterns in Deep Convolutional Networks: Generative Adversarial Phonology and Reduplication