Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm
We present an expansion of FLEET, a machine-learning algorithm optimized to select transients that are most likely tidal disruption events (TDEs). FLEET is based on a random forest algorithm trained on both the light curves and host galaxy information of 4779 spectroscopically classified transients....
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/acc535 |
| _version_ | 1827830190952153088 |
|---|---|
| author | Sebastian Gomez V. Ashley Villar Edo Berger Suvi Gezari Sjoert van Velzen Matt Nicholl Peter K. Blanchard Kate. D. Alexander |
| author_facet | Sebastian Gomez V. Ashley Villar Edo Berger Suvi Gezari Sjoert van Velzen Matt Nicholl Peter K. Blanchard Kate. D. Alexander |
| author_sort | Sebastian Gomez |
| collection | DOAJ |
| description | We present an expansion of FLEET, a machine-learning algorithm optimized to select transients that are most likely tidal disruption events (TDEs). FLEET is based on a random forest algorithm trained on both the light curves and host galaxy information of 4779 spectroscopically classified transients. We find that for transients with a probability of being a TDE, P (TDE) > 0.5, we can successfully recover TDEs with ≈40% completeness and ≈30% purity when using their first 20 days of photometry or a similar completeness and ≈50% purity when including 40 days of photometry, an improvement of almost 2 orders of magnitude compared to random selection. Alternatively, we can recover TDEs with a maximum purity of ≈80% and a completeness of ≈30% when considering only transients with P (TDE) > 0.8. We explore the use of FLEET for future time-domain surveys such as the Legacy Survey of Space and Time on the Vera C. Rubin Observatory (Rubin) and the Nancy Grace Roman Space Telescope (Roman). We estimate that ∼10 ^4 well-observed TDEs could be discovered every year by Rubin and ∼200 TDEs by Roman. Finally, we run FLEET on the TDEs from our Rubin survey simulation and find that we can recover ∼30% of them at redshift z < 0.5 with P (TDE) > 0.5, or ∼3000 TDEs yr ^–1 that FLEET could uncover from the Rubin stream. We have demonstrated that we will be able to run FLEET on Rubin photometry as soon as this survey begins. FLEET is provided as an open source package on GitHub: https://github.com/gmzsebastian/FLEET . |
| first_indexed | 2024-03-12T04:17:29Z |
| format | Article |
| id | doaj.art-df3dbe3942204be9a231104f66bcb1a9 |
| institution | Directory Open Access Journal |
| issn | 1538-4357 |
| language | English |
| last_indexed | 2024-03-12T04:17:29Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj.art-df3dbe3942204be9a231104f66bcb1a92023-09-03T10:35:21ZengIOP PublishingThe Astrophysical Journal1538-43572023-01-01949211310.3847/1538-4357/acc535Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning AlgorithmSebastian Gomez0https://orcid.org/0000-0001-6395-6702V. Ashley Villar1https://orcid.org/0000-0002-5814-4061Edo Berger2https://orcid.org/0000-0002-9392-9681Suvi Gezari3https://orcid.org/0000-0003-3703-5154Sjoert van Velzen4https://orcid.org/0000-0002-3859-8074Matt Nicholl5https://orcid.org/0000-0002-2555-3192Peter K. Blanchard6https://orcid.org/0000-0003-0526-2248Kate. D. Alexander7https://orcid.org/0000-0002-8297-2473Space Telescope Science Institute , 3700 San Martin Drive, Baltimore, MD 21218, USA ; sgomez@stsci.eduDepartment of Astronomy & Astrophysics, The Pennsylvania State University , University Park, PA 16802, USA; Institute for Computational & Data Sciences, The Pennsylvania State University , University Park, PA 16802, USA; Institute for Gravitation and the Cosmos, The Pennsylvania State University , University Park, PA 16802, USACenter for Astrophysics ∣ Harvard & Smithsonian , 60 Garden Street, Cambridge, MA 02138-1516, USA; The NSF AI Institute for Artificial Intelligence and Fundamental Interactions , USASpace Telescope Science Institute , 3700 San Martin Drive, Baltimore, MD 21218, USA ; sgomez@stsci.eduLeiden Observatory, Leiden University , P.O. Box 9513, 2300 RA Leiden, The NetherlandsBirmingham Institute for Gravitational Wave Astronomy and School of Physics and Astronomy, University of Birmingham , Birmingham B15 2TT, UKCenter for Interdisciplinary Exploration and Research in Astrophysics and Department of Physics and Astronomy, Northwestern University , 1800 Sherman Avenue, 8th Floor, Evanston, IL 60201, USACenter for Interdisciplinary Exploration and Research in Astrophysics and Department of Physics and Astronomy, Northwestern University , 1800 Sherman Avenue, 8th Floor, Evanston, IL 60201, USA; Department of Astronomy/Steward Observatory, University of Arizona , 933 North Cherry Avenue, Tucson, AZ 85721-0065, USAWe present an expansion of FLEET, a machine-learning algorithm optimized to select transients that are most likely tidal disruption events (TDEs). FLEET is based on a random forest algorithm trained on both the light curves and host galaxy information of 4779 spectroscopically classified transients. We find that for transients with a probability of being a TDE, P (TDE) > 0.5, we can successfully recover TDEs with ≈40% completeness and ≈30% purity when using their first 20 days of photometry or a similar completeness and ≈50% purity when including 40 days of photometry, an improvement of almost 2 orders of magnitude compared to random selection. Alternatively, we can recover TDEs with a maximum purity of ≈80% and a completeness of ≈30% when considering only transients with P (TDE) > 0.8. We explore the use of FLEET for future time-domain surveys such as the Legacy Survey of Space and Time on the Vera C. Rubin Observatory (Rubin) and the Nancy Grace Roman Space Telescope (Roman). We estimate that ∼10 ^4 well-observed TDEs could be discovered every year by Rubin and ∼200 TDEs by Roman. Finally, we run FLEET on the TDEs from our Rubin survey simulation and find that we can recover ∼30% of them at redshift z < 0.5 with P (TDE) > 0.5, or ∼3000 TDEs yr ^–1 that FLEET could uncover from the Rubin stream. We have demonstrated that we will be able to run FLEET on Rubin photometry as soon as this survey begins. FLEET is provided as an open source package on GitHub: https://github.com/gmzsebastian/FLEET .https://doi.org/10.3847/1538-4357/acc535Black hole physicsSupernovaeSurveys |
| spellingShingle | Sebastian Gomez V. Ashley Villar Edo Berger Suvi Gezari Sjoert van Velzen Matt Nicholl Peter K. Blanchard Kate. D. Alexander Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm The Astrophysical Journal Black hole physics Supernovae Surveys |
| title | Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm |
| title_full | Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm |
| title_fullStr | Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm |
| title_full_unstemmed | Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm |
| title_short | Identifying Tidal Disruption Events with an Expansion of the FLEET Machine-learning Algorithm |
| title_sort | identifying tidal disruption events with an expansion of the fleet machine learning algorithm |
| topic | Black hole physics Supernovae Surveys |
| url | https://doi.org/10.3847/1538-4357/acc535 |
| work_keys_str_mv | AT sebastiangomez identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT vashleyvillar identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT edoberger identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT suvigezari identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT sjoertvanvelzen identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT mattnicholl identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT peterkblanchard identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm AT katedalexander identifyingtidaldisruptioneventswithanexpansionofthefleetmachinelearningalgorithm |