sbi reloaded: a toolkit for simulation-based inference workflows

Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, id...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Boelts, Jan, Deistler, Michael, Gloeckler, Manuel, Tejero-Cantero, Álvaro, Lueckmann, Jan-Matthis, Moss, Guy, Steinbach, Peter, Moreau, Thomas, Muratore, Fabio, Linhart, Julia, Durkan, Conor, Vetter, Julius, Miller, Benjamin Kurt, Herold, Maternus, Ziaeemehr, Abolfazl, Pals, Matthijs, Gruner, Theo, Bischoff, Sebastian, Krouglova, Nastya, Gao, Richard, Lappalainen, Janne K, Mucsányi, Bálint, Pei, Felix, Schulz, Auguste, Stefanidi, Zinovia, Rodrigues, Pedro, Schröder, Cornelius, Zaid, Faried Abu, Beck, Jonas, Kapoor, Jaivardhan, Greenberg, David S, Gonçalves, Pedro J, Macke, Jakob H
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Boelts, Jan
Deistler, Michael
Gloeckler, Manuel
Tejero-Cantero, Álvaro
Lueckmann, Jan-Matthis
Moss, Guy
Steinbach, Peter
Moreau, Thomas
Muratore, Fabio
Linhart, Julia
Durkan, Conor
Vetter, Julius
Miller, Benjamin Kurt
Herold, Maternus
Ziaeemehr, Abolfazl
Pals, Matthijs
Gruner, Theo
Bischoff, Sebastian
Krouglova, Nastya
Gao, Richard
Lappalainen, Janne K
Mucsányi, Bálint
Pei, Felix
Schulz, Auguste
Stefanidi, Zinovia
Rodrigues, Pedro
Schröder, Cornelius
Zaid, Faried Abu
Beck, Jonas
Kapoor, Jaivardhan
Greenberg, David S
Gonçalves, Pedro J
Macke, Jakob H
description Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.
doi_str_mv 10.48550/arxiv.2411.17337
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2411_17337</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2411_17337</sourcerecordid><originalsourceid>FETCH-arxiv_primary_2411_173373</originalsourceid><addsrcrecordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjE01DM0NzY252RwKE7KVChKzclPTElNsVJIVCjJz8_JzixRSMsvUijOzC3NSSzJzM_TTUosTk1RyMxLSy1KzUtOVSjPL8pOy8kvL-ZhYE1LzClO5YXS3Azybq4hzh66YLviC4oycxOLKuNBdsaD7TQmrAIAFbY3Hw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>sbi reloaded: a toolkit for simulation-based inference workflows</title><source>arXiv.org</source><creator>Boelts, Jan ; Deistler, Michael ; Gloeckler, Manuel ; Tejero-Cantero, Álvaro ; Lueckmann, Jan-Matthis ; Moss, Guy ; Steinbach, Peter ; Moreau, Thomas ; Muratore, Fabio ; Linhart, Julia ; Durkan, Conor ; Vetter, Julius ; Miller, Benjamin Kurt ; Herold, Maternus ; Ziaeemehr, Abolfazl ; Pals, Matthijs ; Gruner, Theo ; Bischoff, Sebastian ; Krouglova, Nastya ; Gao, Richard ; Lappalainen, Janne K ; Mucsányi, Bálint ; Pei, Felix ; Schulz, Auguste ; Stefanidi, Zinovia ; Rodrigues, Pedro ; Schröder, Cornelius ; Zaid, Faried Abu ; Beck, Jonas ; Kapoor, Jaivardhan ; Greenberg, David S ; Gonçalves, Pedro J ; Macke, Jakob H</creator><creatorcontrib>Boelts, Jan ; Deistler, Michael ; Gloeckler, Manuel ; Tejero-Cantero, Álvaro ; Lueckmann, Jan-Matthis ; Moss, Guy ; Steinbach, Peter ; Moreau, Thomas ; Muratore, Fabio ; Linhart, Julia ; Durkan, Conor ; Vetter, Julius ; Miller, Benjamin Kurt ; Herold, Maternus ; Ziaeemehr, Abolfazl ; Pals, Matthijs ; Gruner, Theo ; Bischoff, Sebastian ; Krouglova, Nastya ; Gao, Richard ; Lappalainen, Janne K ; Mucsányi, Bálint ; Pei, Felix ; Schulz, Auguste ; Stefanidi, Zinovia ; Rodrigues, Pedro ; Schröder, Cornelius ; Zaid, Faried Abu ; Beck, Jonas ; Kapoor, Jaivardhan ; Greenberg, David S ; Gonçalves, Pedro J ; Macke, Jakob H</creatorcontrib><description>Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.</description><identifier>DOI: 10.48550/arxiv.2411.17337</identifier><language>eng</language><subject>Computer Science - Learning</subject><creationdate>2024-11</creationdate><rights>http://creativecommons.org/licenses/by-sa/4.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2411.17337$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2411.17337$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Boelts, Jan</creatorcontrib><creatorcontrib>Deistler, Michael</creatorcontrib><creatorcontrib>Gloeckler, Manuel</creatorcontrib><creatorcontrib>Tejero-Cantero, Álvaro</creatorcontrib><creatorcontrib>Lueckmann, Jan-Matthis</creatorcontrib><creatorcontrib>Moss, Guy</creatorcontrib><creatorcontrib>Steinbach, Peter</creatorcontrib><creatorcontrib>Moreau, Thomas</creatorcontrib><creatorcontrib>Muratore, Fabio</creatorcontrib><creatorcontrib>Linhart, Julia</creatorcontrib><creatorcontrib>Durkan, Conor</creatorcontrib><creatorcontrib>Vetter, Julius</creatorcontrib><creatorcontrib>Miller, Benjamin Kurt</creatorcontrib><creatorcontrib>Herold, Maternus</creatorcontrib><creatorcontrib>Ziaeemehr, Abolfazl</creatorcontrib><creatorcontrib>Pals, Matthijs</creatorcontrib><creatorcontrib>Gruner, Theo</creatorcontrib><creatorcontrib>Bischoff, Sebastian</creatorcontrib><creatorcontrib>Krouglova, Nastya</creatorcontrib><creatorcontrib>Gao, Richard</creatorcontrib><creatorcontrib>Lappalainen, Janne K</creatorcontrib><creatorcontrib>Mucsányi, Bálint</creatorcontrib><creatorcontrib>Pei, Felix</creatorcontrib><creatorcontrib>Schulz, Auguste</creatorcontrib><creatorcontrib>Stefanidi, Zinovia</creatorcontrib><creatorcontrib>Rodrigues, Pedro</creatorcontrib><creatorcontrib>Schröder, Cornelius</creatorcontrib><creatorcontrib>Zaid, Faried Abu</creatorcontrib><creatorcontrib>Beck, Jonas</creatorcontrib><creatorcontrib>Kapoor, Jaivardhan</creatorcontrib><creatorcontrib>Greenberg, David S</creatorcontrib><creatorcontrib>Gonçalves, Pedro J</creatorcontrib><creatorcontrib>Macke, Jakob H</creatorcontrib><title>sbi reloaded: a toolkit for simulation-based inference workflows</title><description>Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.</description><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpjYJA0NNAzsTA1NdBPLKrILNMzMjE01DM0NzY252RwKE7KVChKzclPTElNsVJIVCjJz8_JzixRSMsvUijOzC3NSSzJzM_TTUosTk1RyMxLSy1KzUtOVSjPL8pOy8kvL-ZhYE1LzClO5YXS3Azybq4hzh66YLviC4oycxOLKuNBdsaD7TQmrAIAFbY3Hw</recordid><startdate>20241126</startdate><enddate>20241126</enddate><creator>Boelts, Jan</creator><creator>Deistler, Michael</creator><creator>Gloeckler, Manuel</creator><creator>Tejero-Cantero, Álvaro</creator><creator>Lueckmann, Jan-Matthis</creator><creator>Moss, Guy</creator><creator>Steinbach, Peter</creator><creator>Moreau, Thomas</creator><creator>Muratore, Fabio</creator><creator>Linhart, Julia</creator><creator>Durkan, Conor</creator><creator>Vetter, Julius</creator><creator>Miller, Benjamin Kurt</creator><creator>Herold, Maternus</creator><creator>Ziaeemehr, Abolfazl</creator><creator>Pals, Matthijs</creator><creator>Gruner, Theo</creator><creator>Bischoff, Sebastian</creator><creator>Krouglova, Nastya</creator><creator>Gao, Richard</creator><creator>Lappalainen, Janne K</creator><creator>Mucsányi, Bálint</creator><creator>Pei, Felix</creator><creator>Schulz, Auguste</creator><creator>Stefanidi, Zinovia</creator><creator>Rodrigues, Pedro</creator><creator>Schröder, Cornelius</creator><creator>Zaid, Faried Abu</creator><creator>Beck, Jonas</creator><creator>Kapoor, Jaivardhan</creator><creator>Greenberg, David S</creator><creator>Gonçalves, Pedro J</creator><creator>Macke, Jakob H</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20241126</creationdate><title>sbi reloaded: a toolkit for simulation-based inference workflows</title><author>Boelts, Jan ; Deistler, Michael ; Gloeckler, Manuel ; Tejero-Cantero, Álvaro ; Lueckmann, Jan-Matthis ; Moss, Guy ; Steinbach, Peter ; Moreau, Thomas ; Muratore, Fabio ; Linhart, Julia ; Durkan, Conor ; Vetter, Julius ; Miller, Benjamin Kurt ; Herold, Maternus ; Ziaeemehr, Abolfazl ; Pals, Matthijs ; Gruner, Theo ; Bischoff, Sebastian ; Krouglova, Nastya ; Gao, Richard ; Lappalainen, Janne K ; Mucsányi, Bálint ; Pei, Felix ; Schulz, Auguste ; Stefanidi, Zinovia ; Rodrigues, Pedro ; Schröder, Cornelius ; Zaid, Faried Abu ; Beck, Jonas ; Kapoor, Jaivardhan ; Greenberg, David S ; Gonçalves, Pedro J ; Macke, Jakob H</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-arxiv_primary_2411_173373</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Boelts, Jan</creatorcontrib><creatorcontrib>Deistler, Michael</creatorcontrib><creatorcontrib>Gloeckler, Manuel</creatorcontrib><creatorcontrib>Tejero-Cantero, Álvaro</creatorcontrib><creatorcontrib>Lueckmann, Jan-Matthis</creatorcontrib><creatorcontrib>Moss, Guy</creatorcontrib><creatorcontrib>Steinbach, Peter</creatorcontrib><creatorcontrib>Moreau, Thomas</creatorcontrib><creatorcontrib>Muratore, Fabio</creatorcontrib><creatorcontrib>Linhart, Julia</creatorcontrib><creatorcontrib>Durkan, Conor</creatorcontrib><creatorcontrib>Vetter, Julius</creatorcontrib><creatorcontrib>Miller, Benjamin Kurt</creatorcontrib><creatorcontrib>Herold, Maternus</creatorcontrib><creatorcontrib>Ziaeemehr, Abolfazl</creatorcontrib><creatorcontrib>Pals, Matthijs</creatorcontrib><creatorcontrib>Gruner, Theo</creatorcontrib><creatorcontrib>Bischoff, Sebastian</creatorcontrib><creatorcontrib>Krouglova, Nastya</creatorcontrib><creatorcontrib>Gao, Richard</creatorcontrib><creatorcontrib>Lappalainen, Janne K</creatorcontrib><creatorcontrib>Mucsányi, Bálint</creatorcontrib><creatorcontrib>Pei, Felix</creatorcontrib><creatorcontrib>Schulz, Auguste</creatorcontrib><creatorcontrib>Stefanidi, Zinovia</creatorcontrib><creatorcontrib>Rodrigues, Pedro</creatorcontrib><creatorcontrib>Schröder, Cornelius</creatorcontrib><creatorcontrib>Zaid, Faried Abu</creatorcontrib><creatorcontrib>Beck, Jonas</creatorcontrib><creatorcontrib>Kapoor, Jaivardhan</creatorcontrib><creatorcontrib>Greenberg, David S</creatorcontrib><creatorcontrib>Gonçalves, Pedro J</creatorcontrib><creatorcontrib>Macke, Jakob H</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Boelts, Jan</au><au>Deistler, Michael</au><au>Gloeckler, Manuel</au><au>Tejero-Cantero, Álvaro</au><au>Lueckmann, Jan-Matthis</au><au>Moss, Guy</au><au>Steinbach, Peter</au><au>Moreau, Thomas</au><au>Muratore, Fabio</au><au>Linhart, Julia</au><au>Durkan, Conor</au><au>Vetter, Julius</au><au>Miller, Benjamin Kurt</au><au>Herold, Maternus</au><au>Ziaeemehr, Abolfazl</au><au>Pals, Matthijs</au><au>Gruner, Theo</au><au>Bischoff, Sebastian</au><au>Krouglova, Nastya</au><au>Gao, Richard</au><au>Lappalainen, Janne K</au><au>Mucsányi, Bálint</au><au>Pei, Felix</au><au>Schulz, Auguste</au><au>Stefanidi, Zinovia</au><au>Rodrigues, Pedro</au><au>Schröder, Cornelius</au><au>Zaid, Faried Abu</au><au>Beck, Jonas</au><au>Kapoor, Jaivardhan</au><au>Greenberg, David S</au><au>Gonçalves, Pedro J</au><au>Macke, Jakob H</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>sbi reloaded: a toolkit for simulation-based inference workflows</atitle><date>2024-11-26</date><risdate>2024</risdate><abstract>Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.</abstract><doi>10.48550/arxiv.2411.17337</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2411.17337
ispartof
issn
language eng
recordid cdi_arxiv_primary_2411_17337
source arXiv.org
subjects Computer Science - Learning
title sbi reloaded: a toolkit for simulation-based inference workflows
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-29T00%3A03%3A45IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=sbi%20reloaded:%20a%20toolkit%20for%20simulation-based%20inference%20workflows&rft.au=Boelts,%20Jan&rft.date=2024-11-26&rft_id=info:doi/10.48550/arxiv.2411.17337&rft_dat=%3Carxiv_GOX%3E2411_17337%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true