DVGAN: Stabilize Wasserstein GAN training for time-domain Gravitational Wave physics

TL;DR

This paper introduces DVGAN, a novel Wasserstein GAN variant with an auxiliary derivative discriminator, to generate realistic time-domain gravitational wave signals and noise, improving training stability and signal quality.

Contribution

The paper presents DVGAN, a new GAN architecture with an auxiliary derivative discriminator that stabilizes training and enhances the realism of simulated gravitational wave signals.

Findings

Discriminating on derivatives stabilizes GAN training on 1D signals.

DVGAN produces smoother, more realistic signals.

Effective in simulating LIGO detector noise events.

Abstract

Simulating time-domain observations of gravitational wave (GW) detector environments will allow for a better understanding of GW sources, augment datasets for GW signal detection and help in characterizing the noise of the detectors, leading to better physics. This paper presents a novel approach to simulating fixed-length time-domain signals using a three-player Wasserstein Generative Adversarial Network (WGAN), called DVGAN, that includes an auxiliary discriminator that discriminates on the derivatives of input signals. An ablation study is used to compare the effects of including adversarial feedback from an auxiliary derivative discriminator with a vanilla two-player WGAN. We show that discriminating on derivatives can stabilize the learning of GAN components on 1D continuous signals during their training phase. This results in smoother generated signals that are less distinguishable from real samples and better capture the distributions of the training data. DVGAN is also used to simulate real transient noise events captured in the advanced LIGO GW detector.