# Deep Learning Detection of Beyond-General-Relativity Deviations in Gravitational-Wave Signals: A Detection-Threshold Study with Real LIGO Noise

arXiv: https://arxiv.org/abs/2609.19416 · HTML: https://mlsanity.com/paper/2609.19416/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 9 references that are in the map.
- First posted on arXiv 16 Sep 2026.

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

- [Real-time gravitational-wave science with neural posterior estimation](https://mlsanity.com/paper/2106.12594/) (2021)
- [Bayesian parameter estimation using conditional variational autoencoders for gravitational-wave astronomy](https://mlsanity.com/paper/1909.06296/) (2019)
- [Convolutional neural networks: a magic bullet for gravitational-wave detection?](https://mlsanity.com/paper/1904.08693/) (2019)
- [Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation with LIGO Data](https://mlsanity.com/paper/1711.07966/) (2017)

## Built on by (most cited first, 1 of 1)

- [Identifiability Limits of Gravitational Wave Phase Deviations: Multiclass Classification with a Multihead Neural Network](https://mlsanity.com/paper/2609.31993/) (2026)

## Abstract (arXiv)

We study machine-learning detection of controlled beyond-General-Relativity (beyond-GR) deviations in gravitational-wave signals, using both synthetic aLIGO-PSD noise and real LIGO H1 detector strain. Three deviation families are applied to General-Relativistic inspiral-merger-ringdown waveforms: amplitude modulation, phase modulation, and frequency modulation, each parameterized by a dimensionless strength coefficient β. A hybrid classifier combining a one-dimensional convolutional neural network with ten hand-crafted waveform statistics is trained on GR and modified waveforms and tested on a deviation type excluded from training. The central result is a quantitative detectability curve as a function of β. Using the real GW150914 strain as a template and real H1 detector noise, we find a detection threshold at β ≈ 0.25, with accuracy rising smoothly from chance at β ≤ 0.2 to perfect classification at β ≥ 0.5. The threshold value is specific to the quadratic-in-time modulation form adopted here and should not be interpreted as a generic constraint on beyond-GR parameters. We nevertheless argue that the negative result at small β is informative: it establishes a quantitative limit on machine-learning-only beyond-GR searches in real detector noise, in the absence of matched-filter signal extraction.