DCTRGAN: Improving the Precision of Generative Models with Reweighting

TL;DR

This paper presents DCTRGAN, a reweighting method that enhances the fidelity of GAN-generated samples, improving their accuracy in simulations without sacrificing statistical power, applicable broadly to generative models.

Contribution

Introduces a post-hoc reweighting correction for deep generative models, improving their accuracy in representing complex probability densities.

Findings

Weighted GAN samples show significant fidelity improvements.

The method maintains statistical power while enhancing accuracy.

Applicable to high energy physics simulations and beyond.

Abstract

Significant advances in deep learning have led to more widely used and precise neural network-based generative models such as Generative Adversarial Networks (GANs). We introduce a post-hoc correction to deep generative models to further improve their fidelity, based on the Deep neural networks using the Classification for Tuning and Reweighting (DCTR) protocol. The correction takes the form of a reweighting function that can be applied to generated examples when making predictions from the simulation. We illustrate this approach using GANs trained on standard multimodal probability densities as well as calorimeter simulations from high energy physics. We show that the weighted GAN examples significantly improve the accuracy of the generated samples without a large loss in statistical power. This approach could be applied to any generative model and is a promising refinement method for high energy physics applications and beyond.