# Training Strategies for Deep Learning Gravitational-Wave Searches

arXiv: https://arxiv.org/abs/2106.03741 · HTML: https://mlsanity.com/paper/2106.03741/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 42 references that are in the map.
- First posted on arXiv 7 Jun 2021.

## Builds on (most cited first, 4 of 4)

- [Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomy](https://mlsanity.com/paper/1909.06296/) (2019)
- [Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation: Results with Advanced LIGO Data](https://mlsanity.com/paper/1711.03121/) (2017)
- [Advances in Machine and Deep Learning for Modeling and Real-time Detection of Multi-Messenger Sources](https://mlsanity.com/paper/2105.06479/) (2021)
- [Generalized Approach to Matched Filtering using Neural Networks](https://mlsanity.com/paper/2104.03961/) (2021)

## Built on by (most cited first, 5 of 5)

- [Inference-optimized AI and high performance computing for gravitational wave detection at scale](https://mlsanity.com/paper/2201.11133/) (2022)
- [AI and extreme scale computing to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non-precessing binary black hole mergers](https://mlsanity.com/paper/2112.07669/) (2021)
- [DeepSNR: A deep learning foundation for offline gravitational wave detection](https://mlsanity.com/paper/2207.04749/) (2022)
- [Learning and Interpreting Gravitational-Wave Features from CNNs with a Random Forest Approach](https://mlsanity.com/paper/2505.20357/) (2025)
- [AI-enabled gravitational-waves searches for binary neutron stars at optimal sensitivity](https://mlsanity.com/paper/2607.01372/) (2026)

## Abstract (arXiv)

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 ≥ 91.5% of the sensitivity of the matched-filter search down to a false-alarm rate of 1 per month.