Training Strategies for Deep Learning Gravitational-Wave Searches

TL;DR

This paper explores training strategies for deep learning in gravitational-wave detection, demonstrating that certain data presentation methods improve sensitivity and that modified algorithms can approach traditional matched-filter performance.

Contribution

It systematically evaluates training data strategies for deep learning in gravitational-wave searches and introduces USR to enhance detection sensitivity.

Findings

Deep learning can generalize low SNR signals to high SNR.

Training with low SNR signals early improves convergence.

Modified USR method retains over 91.5% sensitivity of matched-filter.

Abstract

Compact binary systems emit gravitational radiation which is potentially detectable by current Earth bound detectors. Extracting these signals from the instruments' background noise is a complex problem and the computational cost of most current searches depends on the complexity of the source model. Deep learning may be capable of finding signals where current algorithms hit computational limits. Here we restrict our analysis to signals from non-spinning binary black holes and systematically test different strategies by which training data is presented to the networks. To assess the impact of the training strategies, we re-analyze the first published networks and directly compare them to an equivalent matched-filter search. We find that the deep learning algorithms can generalize low signal-to-noise ratio (SNR) signals to high SNR ones but not vice versa. As such, it is not beneficial to provide high SNR signals during training, and fastest convergence is achieved when low SNR samples are provided early on. During testing we found that the networks are sometimes unable to recover any signals when a false alarm probability $<10^{-3}$ is required. We resolve this restriction by applying a modification we call unbounded Softmax replacement (USR) after training. With this alteration we find that the machine learning search retains $\geq 91.5\%$ of the sensitivity of the matched-filter search down to a false-alarm rate of 1 per month.