Adversarial Learning with Local Coordinate Coding

Abstract

Generative adversarial networks (GANs) aim to generate realistic data from some prior distribution (e.g., Gaussian noises). However, such prior distribution is often independent of real data and thus may lose semantic information (e.g., geometric structure or content in images) of data. In practice, the semantic information might be represented by some latent distribution learned from data, which, however, is hard to be used for sampling in GANs. In this paper, rather than sampling from the pre-defined prior distribution, we propose a Local Coordinate Coding (LCC) based sampling method to improve GANs. We derive a generalization bound for LCC based GANs and prove that a small dimensional input is sufficient to achieve good generalization. Extensive experiments on various real-world datasets demonstrate the effectiveness of the proposed method.

14 pages, 7 figures, 4 tables

References in corpus (2)

• A Note on the Inception Score
• On the Discrimination-Generalization Tradeoff in GANs

Cited by in corpus (6)

• Discrimination-aware Network Pruning for Deep Model Compression
• Closed-loop Matters: Dual Regression Networks for Single Image Super-Resolution
• Generative Low-bitwidth Data Free Quantization
• Location-aware Graph Convolutional Networks for Video Question Answering
• EFANet: Exchangeable Feature Alignment Network for Arbitrary Style Transfer
• Reference Network for Neural Machine Translation