G2R Bound: A Generalization Bound for Supervised Learning from GAN-Synthetic Data

TL;DR

This paper introduces a theoretical generalization bound for supervised learning models trained on GAN-generated synthetic data, addressing accuracy and privacy concerns.

Contribution

It proposes a novel generalization bound that quantifies the gap between training on GAN-synthetic data and testing on real data, aiding model development.

Findings

Provides a bound to measure generalization gap

Helps improve classifier performance with synthetic data

Addresses privacy-preserving data sharing scenarios

Abstract

Performing supervised learning from the data synthesized by using Generative Adversarial Networks (GANs), dubbed GAN-synthetic data, has two important applications. First, GANs may generate more labeled training data, which may help improve classification accuracy. Second, in scenarios where real data cannot be released outside certain premises for privacy and/or security reasons, using GAN- synthetic data to conduct training is a plausible alternative. This paper proposes a generalization bound to guarantee the generalization capability of a classifier learning from GAN-synthetic data. This generalization bound helps developers gauge the generalization gap between learning from synthetic data and testing on real data, and can therefore provide the clues to improve the generalization capability.