Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories

Machine learning pipelines for astrochemical inventories have been introduced as a useful addition to the astrochemist toolbox, having first been used to model and predict column densities in the Taurus Molecular Cloud (TMC-1). Rapid changes in the field of machine learning have provided new tools i...

Full description

Saved in:
Bibliographic Details
Main Authors: Hannah Toru Shay, Haley N. Scolati, Gabi Wenzel, Kin Long Kelvin Lee, Aravindh N. Marimuthu, Brett A. McGuire
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astrophysical Journal
Subjects:
Online Access:https://doi.org/10.3847/1538-4357/adc80b
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850150675182256128
author Hannah Toru Shay
Haley N. Scolati
Gabi Wenzel
Kin Long Kelvin Lee
Aravindh N. Marimuthu
Brett A. McGuire
author_facet Hannah Toru Shay
Haley N. Scolati
Gabi Wenzel
Kin Long Kelvin Lee
Aravindh N. Marimuthu
Brett A. McGuire
author_sort Hannah Toru Shay
collection DOAJ
description Machine learning pipelines for astrochemical inventories have been introduced as a useful addition to the astrochemist toolbox, having first been used to model and predict column densities in the Taurus Molecular Cloud (TMC-1). Rapid changes in the field of machine learning have provided new tools in optimizing this pipeline, including improved vector representations. Furthermore, the addition of new detections since the original model allows for a retrospective analysis of model performance, in addition to new data for the model. This study revisits TMC-1, investigating both effects of an increased detection inventory on the model and changes to various portions of the pipeline, yielding a significant improvement in column density predictions. Through these comparisons, we attempt to derive insight into the ultimate effectiveness of these models, as well as their current limitations and words of caution in their use and interpretation. Finally, we provide suggestions for future machine learning of interstellar sources.
format Article
id doaj-art-87f0e1506de143deb260fda6eee17143
institution OA Journals
issn 1538-4357
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal
spelling doaj-art-87f0e1506de143deb260fda6eee171432025-08-20T02:26:28ZengIOP PublishingThe Astrophysical Journal1538-43572025-01-01985112310.3847/1538-4357/adc80bExploring Effects of Modified Machine Learning Pipelines of Astrochemical InventoriesHannah Toru Shay0https://orcid.org/0000-0002-8721-0988Haley N. Scolati1https://orcid.org/0000-0002-8505-4459Gabi Wenzel2https://orcid.org/0000-0002-0332-2641Kin Long Kelvin Lee3https://orcid.org/0000-0002-1903-9242Aravindh N. Marimuthu4https://orcid.org/0000-0001-5444-6401Brett A. McGuire5https://orcid.org/0000-0003-1254-4817Department of Chemistry, Massachusetts Institute of Technology , Cambridge, MA 02139, USA ; hannahts@mit.edu, brettmc@mit.eduNational Radio Astronomy Observatory , Charlottesville, VA 22903, USA; Department of Chemistry, University of Virginia , Charlottesville, VA 22903, USADepartment of Chemistry, Massachusetts Institute of Technology , Cambridge, MA 02139, USA ; hannahts@mit.edu, brettmc@mit.eduIntel Labs , Intel Corporation, 2111 NE 25th Ave., Hillsboro, OR 97124, USADepartment of Chemistry, Massachusetts Institute of Technology , Cambridge, MA 02139, USA ; hannahts@mit.edu, brettmc@mit.eduDepartment of Chemistry, Massachusetts Institute of Technology , Cambridge, MA 02139, USA ; hannahts@mit.edu, brettmc@mit.edu; National Radio Astronomy Observatory , Charlottesville, VA 22903, USAMachine learning pipelines for astrochemical inventories have been introduced as a useful addition to the astrochemist toolbox, having first been used to model and predict column densities in the Taurus Molecular Cloud (TMC-1). Rapid changes in the field of machine learning have provided new tools in optimizing this pipeline, including improved vector representations. Furthermore, the addition of new detections since the original model allows for a retrospective analysis of model performance, in addition to new data for the model. This study revisits TMC-1, investigating both effects of an increased detection inventory on the model and changes to various portions of the pipeline, yielding a significant improvement in column density predictions. Through these comparisons, we attempt to derive insight into the ultimate effectiveness of these models, as well as their current limitations and words of caution in their use and interpretation. Finally, we provide suggestions for future machine learning of interstellar sources.https://doi.org/10.3847/1538-4357/adc80bAstrochemistryInterdisciplinary astronomyInterstellar molecules
spellingShingle Hannah Toru Shay
Haley N. Scolati
Gabi Wenzel
Kin Long Kelvin Lee
Aravindh N. Marimuthu
Brett A. McGuire
Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
The Astrophysical Journal
Astrochemistry
Interdisciplinary astronomy
Interstellar molecules
title Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
title_full Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
title_fullStr Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
title_full_unstemmed Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
title_short Exploring Effects of Modified Machine Learning Pipelines of Astrochemical Inventories
title_sort exploring effects of modified machine learning pipelines of astrochemical inventories
topic Astrochemistry
Interdisciplinary astronomy
Interstellar molecules
url https://doi.org/10.3847/1538-4357/adc80b
work_keys_str_mv AT hannahtorushay exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories
AT haleynscolati exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories
AT gabiwenzel exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories
AT kinlongkelvinlee exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories
AT aravindhnmarimuthu exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories
AT brettamcguire exploringeffectsofmodifiedmachinelearningpipelinesofastrochemicalinventories