Evaluating machine learning models for supernova gravitational wave signal classification
We investigate the potential of using gravitational wave (GW) signals from rotating core-collapse supernovae to probe the equation of state (EOS) of nuclear matter. By generating GW signals from simulations with various EOSs, we train machine learning models to classify them and evaluate their perfo...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | Machine Learning: Science and Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2632-2153/ada33a |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841558780182003712 |
|---|---|
| author | Y Sultan Abylkairov Matthew C Edwards Daniil Orel Ayan Mitra Bekdaulet Shukirgaliyev Ernazar Abdikamalov |
| author_facet | Y Sultan Abylkairov Matthew C Edwards Daniil Orel Ayan Mitra Bekdaulet Shukirgaliyev Ernazar Abdikamalov |
| author_sort | Y Sultan Abylkairov |
| collection | DOAJ |
| description | We investigate the potential of using gravitational wave (GW) signals from rotating core-collapse supernovae to probe the equation of state (EOS) of nuclear matter. By generating GW signals from simulations with various EOSs, we train machine learning models to classify them and evaluate their performance. Our study builds on previous work by examining how different machine learning models, parameters, and data preprocessing techniques impact classification accuracy. We test convolutional and recurrent neural networks, as well as six classical algorithms: random forest, support vector machines, naïve Bayes(NB), logistic regression, k -nearest neighbors, and eXtreme gradient boosting. All models, except NB, achieve over 90 per cent accuracy on our dataset. Additionally, we assess the impact of approximating the GW signal using the general relativistic effective potential (GREP) on EOS classification. We find that models trained on GREP data exhibit low classification accuracy. However, normalizing time by the peak signal frequency, which partially compensates for the absence of the time dilation effect in GREP, leads to a notable improvement in accuracy. Despite this, the accuracy does not exceed 70 per cent, suggesting that GREP lacks the precision necessary for EOS classification. Finally, our study has several limitations, including the omission of detector noise and the focus on a single progenitor mass model, which will be addressed in future works. |
| format | Article |
| id | doaj-art-8cb1df71587d455ab1cbbf1a8f58636e |
| institution | Kabale University |
| issn | 2632-2153 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Machine Learning: Science and Technology |
| spelling | doaj-art-8cb1df71587d455ab1cbbf1a8f58636e2025-01-06T05:13:59ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-015404507710.1088/2632-2153/ada33aEvaluating machine learning models for supernova gravitational wave signal classificationY Sultan Abylkairov0https://orcid.org/0000-0003-2656-7294Matthew C Edwards1Daniil Orel2Ayan Mitra3Bekdaulet Shukirgaliyev4https://orcid.org/0000-0002-4601-7065Ernazar Abdikamalov5Department of Mathematics, Nazarbayev University , 010000 Astana, Kazakhstan; Energetic Cosmos Laboratory, Nazarbayev University , 010000 Astana, KazakhstanDepartment of Statistics, University of Auckland , Auckland 1010, New ZealandDepartment of Natural Language Processing, Mohamed bin Zayed University of Artificial Intelligence , Abu Dhabi 54115, United Arab EmiratesCenter for Astrophysical Surveys, National Center for Supercomputing Applications, University of Illinois Urbana-Champaign, Urbana, IL 61801, United States of America; Department of Astronomy, University of Illinois at Urbana-Champaign , Urbana, IL 61801, United States of AmericaHeriot-Watt University Aktobe Campus , 030000 Aktobe, Kazakhstan; K. Zhubanov Aktobe Regional University , 030000 Aktobe, Kazakhstan; Department of Physics, Nazarbayev University , 010000 Astana, KazakhstanEnergetic Cosmos Laboratory, Nazarbayev University , 010000 Astana, Kazakhstan; Department of Physics, Nazarbayev University , 010000 Astana, KazakhstanWe investigate the potential of using gravitational wave (GW) signals from rotating core-collapse supernovae to probe the equation of state (EOS) of nuclear matter. By generating GW signals from simulations with various EOSs, we train machine learning models to classify them and evaluate their performance. Our study builds on previous work by examining how different machine learning models, parameters, and data preprocessing techniques impact classification accuracy. We test convolutional and recurrent neural networks, as well as six classical algorithms: random forest, support vector machines, naïve Bayes(NB), logistic regression, k -nearest neighbors, and eXtreme gradient boosting. All models, except NB, achieve over 90 per cent accuracy on our dataset. Additionally, we assess the impact of approximating the GW signal using the general relativistic effective potential (GREP) on EOS classification. We find that models trained on GREP data exhibit low classification accuracy. However, normalizing time by the peak signal frequency, which partially compensates for the absence of the time dilation effect in GREP, leads to a notable improvement in accuracy. Despite this, the accuracy does not exceed 70 per cent, suggesting that GREP lacks the precision necessary for EOS classification. Finally, our study has several limitations, including the omission of detector noise and the focus on a single progenitor mass model, which will be addressed in future works.https://doi.org/10.1088/2632-2153/ada33agravitational wavesmachine learningdeep learningdata analysissupernovae |
| spellingShingle | Y Sultan Abylkairov Matthew C Edwards Daniil Orel Ayan Mitra Bekdaulet Shukirgaliyev Ernazar Abdikamalov Evaluating machine learning models for supernova gravitational wave signal classification Machine Learning: Science and Technology gravitational waves machine learning deep learning data analysis supernovae |
| title | Evaluating machine learning models for supernova gravitational wave signal classification |
| title_full | Evaluating machine learning models for supernova gravitational wave signal classification |
| title_fullStr | Evaluating machine learning models for supernova gravitational wave signal classification |
| title_full_unstemmed | Evaluating machine learning models for supernova gravitational wave signal classification |
| title_short | Evaluating machine learning models for supernova gravitational wave signal classification |
| title_sort | evaluating machine learning models for supernova gravitational wave signal classification |
| topic | gravitational waves machine learning deep learning data analysis supernovae |
| url | https://doi.org/10.1088/2632-2153/ada33a |
| work_keys_str_mv | AT ysultanabylkairov evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification AT matthewcedwards evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification AT daniilorel evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification AT ayanmitra evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification AT bekdauletshukirgaliyev evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification AT ernazarabdikamalov evaluatingmachinelearningmodelsforsupernovagravitationalwavesignalclassification |