Single-level Adversarial Data Synthesis based on Neural Tangent Kernels

TL;DR

This paper introduces GA-NTK, a single-level adversarial data synthesis model using neural tangent kernels, which simplifies training compared to GANs and provides theoretical convergence insights.

Contribution

Proposes GA-NTK, a novel single-level adversarial generative model based on neural tangent kernels, avoiding GAN training issues and enabling theoretical analysis.

Findings

GA-NTK achieves competitive data synthesis quality.

Theoretical convergence conditions are established.

GA-NTK offers a more stable training process.

Abstract

Abstract Generative adversarial networks (GANs) have achieved impressive performance in data synthesis and have driven the development of many applications. However, GANs are known to be hard to train due to their bilevel objective, which leads to the problems of convergence, mode collapse, and gradient vanishing. In this paper, we propose a new generative model called the generative adversarial NTK (GA-NTK) that has a single-level objective. The GA-NTK keeps the spirit of adversarial learning (which helps generate plausible data) while avoiding the training difficulties of GANs. This is done by modeling the discriminator as a Gaussian process with a neural tangent kernel (NTK-GP) whose training dynamics can be completely described by a closed-form formula. We analyze the convergence behavior of GA-NTK trained by gradient descent and give some sufficient conditions for convergence. We also conduct extensive experiments to study the advantages and limitations of GA-NTK and propose some techniques that make GA-NTK more practical.