Semantic Preserving Generative Adversarial Models

TL;DR

This paper presents a novel generative adversarial model that uses a calibrated classifier instead of a discriminator to generate data with guaranteed semantic properties, applicable across diverse domains.

Contribution

It introduces a new GAN framework replacing the discriminator with a non-differentiable classifier that enforces semantic constraints, improving data generation quality and efficiency.

Findings

Requires less data than traditional GANs

Provides natural stopping criteria for training

Uncovers important data properties and enhances transfer learning

Abstract

We introduce generative adversarial models in which the discriminator is replaced by a calibrated (non-differentiable) classifier repeatedly enhanced by domain relevant features. The role of the classifier is to prove that the actual and generated data differ over a controlled semantic space. We demonstrate that such models have the ability to generate objects with strong guarantees on their properties in a wide range of domains. They require less data than ordinary GANs, provide natural stopping conditions, uncover important properties of the data, and enhance transfer learning. Our techniques can be combined with standard generative models. We demonstrate the usefulness of our approach by applying it to several unrelated domains: generating good locations for cellular antennae, molecule generation preserving key chemical properties, and generating and extrapolating lines from very few data points. Intriguing open problems are presented as well.