Training Strategies for Deep Learning Gravitational-Wave Searches

• URL: http://arxiv.org/abs/2106.03741v1
• Date: Mon, 7 Jun 2021 16:04:29 GMT
• Title: Training Strategies for Deep Learning Gravitational-Wave Searches
• Authors: Marlin B. Sch\"afer (1 and 2), Ond\v{r}ej Zelenka (3 and 4), Alexander H. Nitz (1 and 2), Frank Ohme (1 and 2), Bernd Br\"ugmann (3 and 4) ((1) Max-Planck-Institut f\"ur Gravitationsphysik (Albert-Einstein-Institut), (2) Leibniz Universit\"at Hannover, (3) Friedrich-Schiller-Universit\"at Jena, (4) Michael Stifel Center Jena)
• Abstract summary: We restrict our analysis to signals from non-spinning binary black holes. We systematically test different strategies by which training data is presented to the networks. We find that the deep learning algorithms can generalize low signal-to-noise ratio (SNR) signals to high SNR ones but not vice versa.
• Score: 43.55994393060723
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• 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 $\geq 97.5\%$ of the sensitivity of the matched-filter search down to a false-alarm rate of 1 per month.

Related papers

• A Deep Neural Network Based Reverse Radio Spectrogram Search Algorithm [0.0]
  We present a fast and modular deep learning algorithm to search for lookalike signals of interest in radio spectrogram data. We show that the algorithm retrieves signals with similar appearance, given only the original radio spectrogram data.
  arXiv  Detail & Related papers  (2023-02-24T04:28:46Z)
• MLGWSC-1: The first Machine Learning Gravitational-Wave Search Mock Data Challenge [110.7678032481059]
  We present the results of the first Machine Learning Gravitational-Wave Search Mock Data Challenge (MLGWSC-1). For this challenge, participating groups had to identify gravitational-wave signals from binary black hole mergers of increasing complexity and duration embedded in progressively more realistic noise. Our results show that current machine learning search algorithms may already be sensitive enough in limited parameter regions to be useful for some production settings.
  arXiv  Detail & Related papers  (2022-09-22T16:44:59Z)
• Space-based gravitational wave signal detection and extraction with deep neural network [13.176946557548042]
  Space-based gravitational wave (GW) detectors will be able to observe signals from sources that are otherwise nearly impossible from current ground-based detection. Here, we develop a high-accuracy GW signal detection and extraction method for all space-based GW sources.
  arXiv  Detail & Related papers  (2022-07-15T11:48:15Z)
• DeepSNR: A deep learning foundation for offline gravitational wave detection [0.0]
  This paper introduces the Deep Learning Signal-to-Noise Ratio (DeepSNR) detection pipeline, which uses a novel method for generating a signal-to-noise ratio ranking statistic from deep learning classifiers. The performance of DeepSNR is demonstrated by identifying binary black hole merger candidates versus noise sources in open LIGO data from the first observation run. The results pave the way for DeepSNR to be used in the scientific discovery of gravitational waves and rare signals in broader contexts.
  arXiv  Detail & Related papers  (2022-07-11T10:18:33Z)
• From One to Many: A Deep Learning Coincident Gravitational-Wave Search [58.720142291102135]
  We construct a two-detector search for gravitational waves from binary black hole mergers using neural networks trained on non-spinning binary black hole data from a single detector. We find that none of these simple two-detector networks are capable of improving the sensitivity over applying networks individually to the data from the detectors.
  arXiv  Detail & Related papers  (2021-08-24T13:25:02Z)
• Deep Networks for Direction-of-Arrival Estimation in Low SNR [89.45026632977456]
  We introduce a Convolutional Neural Network (CNN) that is trained from mutli-channel data of the true array manifold matrix. We train a CNN in the low-SNR regime to predict DoAs across all SNRs. Our robust solution can be applied in several fields, ranging from wireless array sensors to acoustic microphones or sonars.
  arXiv  Detail & Related papers  (2020-11-17T12:52:18Z)
• Deep learning for gravitational-wave data analysis: A resampling white-box approach [62.997667081978825]
  We apply Convolutional Neural Networks (CNNs) to detect gravitational wave (GW) signals of compact binary coalescences, using single-interferometer data from LIGO detectors. CNNs were quite precise to detect noise but not sensitive enough to recall GW signals, meaning that CNNs are better for noise reduction than generation of GW triggers.
  arXiv  Detail & Related papers  (2020-09-09T03:28:57Z)
• Detection of gravitational-wave signals from binary neutron star mergers using machine learning [52.77024349608834]
  We introduce a novel neural-network based machine learning algorithm that uses time series strain data from gravitational-wave detectors. We find an improvement by a factor of 6 in sensitivity to signals with signal-to-noise ratio below 25. A conservative estimate indicates that our algorithm introduces on average 10.2 s of latency between signal arrival and generating an alert.
  arXiv  Detail & Related papers  (2020-06-02T10:20:11Z)
• Efficient Training of Deep Classifiers for Wireless Source Identification using Test SNR Estimates [4.44483539967295]
  We study efficient deep learning training algorithms that process wireless signals if a test Signal to Noise Ratio (SNR) estimate is available. For benchmarking, we rely on recent literature on testing deep learning algorithms against two well-known datasets. An erroneous test SNR estimate with a small positive offset is better for training than another having the same error magnitude with a negative offset.
  arXiv  Detail & Related papers  (2019-12-26T16:49:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.