LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology. Ho, M., Bartlett, D. J., Chartier, N., Cuesta-Lazaro, C., Ding, S., Lapel, A., Lemos, P., Lovell, C. C., Makinen, T. L., Modi, C., Pandya, V., Pandey, S., Perez, L. A., Wandelt, B., & Bryan, G. L. The Open Journal of Astrophysics, 7:54, July, 2024. ADS Bibcode: 2024OJAp....7E..54HPaper doi abstract bibtex This paper presents the Learning the Universe Implicit Likelihood Inference (LtU-ILI) pipeline, a codebase for rapid, user-friendly, and cutting-edge machine learning (ML) inference in astrophysics and cosmology. The pipeline includes software for implementing various neural architectures, training schema, priors, and density estimators in a manner easily adaptable to any research workflow. It includes comprehensive validation metrics to assess posterior estimate coverage, enhancing the reliability of inferred results. Additionally, the pipeline is easily parallelizable, designed for efficient exploration of modeling hyperparameters. To demonstrate its capabilities, we present real applications across a range of astrophysics and cosmology problems, such as: estimating galaxy cluster masses from X-ray photometry; inferring cosmology from matter power spectra and halo point clouds; characterising progenitors in gravitational wave signals; capturing physical dust parameters from galaxy colors and luminosities; and establishing properties of semi-analytic models of galaxy formation. We also include exhaustive benchmarking and comparisons of all implemented methods as well as discussions about the challenges and pitfalls of ML inference in astronomical sciences. All code and examples are made publicly available at https://github.com/maho3/ltu-ili.
@article{ho_ltu-ili_2024,
title = {{LtU}-{ILI}: {An} {All}-in-{One} {Framework} for {Implicit} {Inference} in {Astrophysics} and {Cosmology}},
volume = {7},
issn = {2565-6120},
shorttitle = {{LtU}-{ILI}},
url = {https://ui.adsabs.harvard.edu/abs/2024OJAp....7E..54H},
doi = {10.33232/001c.120559},
abstract = {This paper presents the Learning the Universe Implicit Likelihood Inference (LtU-ILI) pipeline, a codebase for rapid, user-friendly, and cutting-edge machine learning (ML) inference in astrophysics and cosmology. The pipeline includes software for implementing various neural architectures, training schema, priors, and density estimators in a manner easily adaptable to any research workflow. It includes comprehensive validation metrics to assess posterior estimate coverage, enhancing the reliability of inferred results. Additionally, the pipeline is easily parallelizable, designed for efficient exploration of modeling hyperparameters. To demonstrate its capabilities, we present real applications across a range of astrophysics and cosmology problems, such as: estimating galaxy cluster masses from X-ray photometry; inferring cosmology from matter power spectra and halo point clouds; characterising progenitors in gravitational wave signals; capturing physical dust parameters from galaxy colors and luminosities; and establishing properties of semi-analytic models of galaxy formation. We also include exhaustive benchmarking and comparisons of all implemented methods as well as discussions about the challenges and pitfalls of ML inference in astronomical sciences. All code and examples are made publicly available at https://github.com/maho3/ltu-ili.},
urldate = {2024-09-03},
journal = {The Open Journal of Astrophysics},
author = {Ho, Matthew and Bartlett, Deaglan J. and Chartier, Nicolas and Cuesta-Lazaro, Carolina and Ding, Simon and Lapel, Axel and Lemos, Pablo and Lovell, Christopher C. and Makinen, T. Lucas and Modi, Chirag and Pandya, Viraj and Pandey, Shivam and Perez, Lucia A. and Wandelt, Benjamin and Bryan, Greg L.},
month = jul,
year = {2024},
note = {ADS Bibcode: 2024OJAp....7E..54H},
keywords = {Astrophysics - Astrophysics of Galaxies, Astrophysics - Cosmology and Nongalactic Astrophysics, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Machine Learning},
pages = {54},
}
Downloads: 0
{"_id":"Eut5s8ReG4XyRBx82","bibbaseid":"ho-bartlett-chartier-cuestalazaro-ding-lapel-lemos-lovell-etal-ltuilianallinoneframeworkforimplicitinferenceinastrophysicsandcosmology-2024","author_short":["Ho, M.","Bartlett, D. J.","Chartier, N.","Cuesta-Lazaro, C.","Ding, S.","Lapel, A.","Lemos, P.","Lovell, C. C.","Makinen, T. L.","Modi, C.","Pandya, V.","Pandey, S.","Perez, L. A.","Wandelt, B.","Bryan, G. L."],"bibdata":{"bibtype":"article","type":"article","title":"LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology","volume":"7","issn":"2565-6120","shorttitle":"LtU-ILI","url":"https://ui.adsabs.harvard.edu/abs/2024OJAp....7E..54H","doi":"10.33232/001c.120559","abstract":"This paper presents the Learning the Universe Implicit Likelihood Inference (LtU-ILI) pipeline, a codebase for rapid, user-friendly, and cutting-edge machine learning (ML) inference in astrophysics and cosmology. The pipeline includes software for implementing various neural architectures, training schema, priors, and density estimators in a manner easily adaptable to any research workflow. It includes comprehensive validation metrics to assess posterior estimate coverage, enhancing the reliability of inferred results. Additionally, the pipeline is easily parallelizable, designed for efficient exploration of modeling hyperparameters. To demonstrate its capabilities, we present real applications across a range of astrophysics and cosmology problems, such as: estimating galaxy cluster masses from X-ray photometry; inferring cosmology from matter power spectra and halo point clouds; characterising progenitors in gravitational wave signals; capturing physical dust parameters from galaxy colors and luminosities; and establishing properties of semi-analytic models of galaxy formation. We also include exhaustive benchmarking and comparisons of all implemented methods as well as discussions about the challenges and pitfalls of ML inference in astronomical sciences. All code and examples are made publicly available at https://github.com/maho3/ltu-ili.","urldate":"2024-09-03","journal":"The Open Journal of Astrophysics","author":[{"propositions":[],"lastnames":["Ho"],"firstnames":["Matthew"],"suffixes":[]},{"propositions":[],"lastnames":["Bartlett"],"firstnames":["Deaglan","J."],"suffixes":[]},{"propositions":[],"lastnames":["Chartier"],"firstnames":["Nicolas"],"suffixes":[]},{"propositions":[],"lastnames":["Cuesta-Lazaro"],"firstnames":["Carolina"],"suffixes":[]},{"propositions":[],"lastnames":["Ding"],"firstnames":["Simon"],"suffixes":[]},{"propositions":[],"lastnames":["Lapel"],"firstnames":["Axel"],"suffixes":[]},{"propositions":[],"lastnames":["Lemos"],"firstnames":["Pablo"],"suffixes":[]},{"propositions":[],"lastnames":["Lovell"],"firstnames":["Christopher","C."],"suffixes":[]},{"propositions":[],"lastnames":["Makinen"],"firstnames":["T.","Lucas"],"suffixes":[]},{"propositions":[],"lastnames":["Modi"],"firstnames":["Chirag"],"suffixes":[]},{"propositions":[],"lastnames":["Pandya"],"firstnames":["Viraj"],"suffixes":[]},{"propositions":[],"lastnames":["Pandey"],"firstnames":["Shivam"],"suffixes":[]},{"propositions":[],"lastnames":["Perez"],"firstnames":["Lucia","A."],"suffixes":[]},{"propositions":[],"lastnames":["Wandelt"],"firstnames":["Benjamin"],"suffixes":[]},{"propositions":[],"lastnames":["Bryan"],"firstnames":["Greg","L."],"suffixes":[]}],"month":"July","year":"2024","note":"ADS Bibcode: 2024OJAp....7E..54H","keywords":"Astrophysics - Astrophysics of Galaxies, Astrophysics - Cosmology and Nongalactic Astrophysics, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Machine Learning","pages":"54","bibtex":"@article{ho_ltu-ili_2024,\n\ttitle = {{LtU}-{ILI}: {An} {All}-in-{One} {Framework} for {Implicit} {Inference} in {Astrophysics} and {Cosmology}},\n\tvolume = {7},\n\tissn = {2565-6120},\n\tshorttitle = {{LtU}-{ILI}},\n\turl = {https://ui.adsabs.harvard.edu/abs/2024OJAp....7E..54H},\n\tdoi = {10.33232/001c.120559},\n\tabstract = {This paper presents the Learning the Universe Implicit Likelihood Inference (LtU-ILI) pipeline, a codebase for rapid, user-friendly, and cutting-edge machine learning (ML) inference in astrophysics and cosmology. The pipeline includes software for implementing various neural architectures, training schema, priors, and density estimators in a manner easily adaptable to any research workflow. It includes comprehensive validation metrics to assess posterior estimate coverage, enhancing the reliability of inferred results. Additionally, the pipeline is easily parallelizable, designed for efficient exploration of modeling hyperparameters. To demonstrate its capabilities, we present real applications across a range of astrophysics and cosmology problems, such as: estimating galaxy cluster masses from X-ray photometry; inferring cosmology from matter power spectra and halo point clouds; characterising progenitors in gravitational wave signals; capturing physical dust parameters from galaxy colors and luminosities; and establishing properties of semi-analytic models of galaxy formation. We also include exhaustive benchmarking and comparisons of all implemented methods as well as discussions about the challenges and pitfalls of ML inference in astronomical sciences. All code and examples are made publicly available at https://github.com/maho3/ltu-ili.},\n\turldate = {2024-09-03},\n\tjournal = {The Open Journal of Astrophysics},\n\tauthor = {Ho, Matthew and Bartlett, Deaglan J. and Chartier, Nicolas and Cuesta-Lazaro, Carolina and Ding, Simon and Lapel, Axel and Lemos, Pablo and Lovell, Christopher C. and Makinen, T. Lucas and Modi, Chirag and Pandya, Viraj and Pandey, Shivam and Perez, Lucia A. and Wandelt, Benjamin and Bryan, Greg L.},\n\tmonth = jul,\n\tyear = {2024},\n\tnote = {ADS Bibcode: 2024OJAp....7E..54H},\n\tkeywords = {Astrophysics - Astrophysics of Galaxies, Astrophysics - Cosmology and Nongalactic Astrophysics, Astrophysics - Instrumentation and Methods for Astrophysics, Computer Science - Machine Learning},\n\tpages = {54},\n}\n\n","author_short":["Ho, M.","Bartlett, D. J.","Chartier, N.","Cuesta-Lazaro, C.","Ding, S.","Lapel, A.","Lemos, P.","Lovell, C. C.","Makinen, T. L.","Modi, C.","Pandya, V.","Pandey, S.","Perez, L. A.","Wandelt, B.","Bryan, G. L."],"key":"ho_ltu-ili_2024","id":"ho_ltu-ili_2024","bibbaseid":"ho-bartlett-chartier-cuestalazaro-ding-lapel-lemos-lovell-etal-ltuilianallinoneframeworkforimplicitinferenceinastrophysicsandcosmology-2024","role":"author","urls":{"Paper":"https://ui.adsabs.harvard.edu/abs/2024OJAp....7E..54H"},"keyword":["Astrophysics - Astrophysics of Galaxies","Astrophysics - Cosmology and Nongalactic Astrophysics","Astrophysics - Instrumentation and Methods for Astrophysics","Computer Science - Machine Learning"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://api.zotero.org/users/2650003/collections/2GYAYTB9/items?key=HjJ9T6LSEhBUjY15bMA811su&format=bibtex&limit=100","dataSources":["CxJmmGJKTkqa9B6aM"],"keywords":["astrophysics - astrophysics of galaxies","astrophysics - cosmology and nongalactic astrophysics","astrophysics - instrumentation and methods for astrophysics","computer science - machine learning"],"search_terms":["ltu","ili","one","framework","implicit","inference","astrophysics","cosmology","ho","bartlett","chartier","cuesta-lazaro","ding","lapel","lemos","lovell","makinen","modi","pandya","pandey","perez","wandelt","bryan"],"title":"LtU-ILI: An All-in-One Framework for Implicit Inference in Astrophysics and Cosmology","year":2024}