A Unified View of cGANs with and without Classifiers

by   Si-An Chen, et al.

Conditional Generative Adversarial Networks (cGANs) are implicit generative models which allow to sample from class-conditional distributions. Existing cGANs are based on a wide range of different discriminator designs and training objectives. One popular design in earlier works is to include a classifier during training with the assumption that good classifiers can help eliminate samples generated with wrong classes. Nevertheless, including classifiers in cGANs often comes with a side effect of only generating easy-to-classify samples. Recently, some representative cGANs avoid the shortcoming and reach state-of-the-art performance without having classifiers. Somehow it remains unanswered whether the classifiers can be resurrected to design better cGANs. In this work, we demonstrate that classifiers can be properly leveraged to improve cGANs. We start by using the decomposition of the joint probability distribution to connect the goals of cGANs and classification as a unified framework. The framework, along with a classic energy model to parameterize distributions, justifies the use of classifiers for cGANs in a principled manner. It explains several popular cGAN variants, such as ACGAN, ProjGAN, and ContraGAN, as special cases with different levels of approximations, which provides a unified view and brings new insights to understanding cGANs. Experimental results demonstrate that the design inspired by the proposed framework outperforms state-of-the-art cGANs on multiple benchmark datasets, especially on the most challenging ImageNet. The code is available at https://github.com/sian-chen/PyTorch-ECGAN.


page 17

page 18

page 19


EVAGAN: Evasion Generative Adversarial Network for Low Data Regimes

A myriad of recent literary works has leveraged generative adversarial n...

Understanding the Intrinsic Robustness of Image Distributions using Conditional Generative Models

Starting with Gilmer et al. (2018), several works have demonstrated the ...

Learning and Evaluating Representations for Deep One-class Classification

We present a two-stage framework for deep one-class classification. We f...

Improving GAN with neighbors embedding and gradient matching

We propose two new techniques for training Generative Adversarial Networ...

Leaving Reality to Imagination: Robust Classification via Generated Datasets

Recent research on robustness has revealed significant performance gaps ...

CoNe: Contrast Your Neighbours for Supervised Image Classification

Image classification is a longstanding problem in computer vision and ma...

Isolation and Impartial Aggregation: A Paradigm of Incremental Learning without Interference

This paper focuses on the prevalent performance imbalance in the stages ...

Please sign up or login with your details

Forgot password? Click here to reset