{"id":"https://openalex.org/W2970936655","doi":"https://doi.org/10.3934/fods.2019019","title":"Partitioned integrators for thermodynamic parameterization of neural networks","display_name":"Partitioned integrators for thermodynamic parameterization of neural networks","publication_year":2019,"publication_date":"2019-01-01","ids":{"openalex":"https://openalex.org/W2970936655","doi":"https://doi.org/10.3934/fods.2019019","mag":"2970936655"},"language":"en","primary_location":{"id":"doi:10.3934/fods.2019019","doi":"https://doi.org/10.3934/fods.2019019","is_oa":true,"landing_page_url":"https://doi.org/10.3934/fods.2019019","pdf_url":null,"source":{"id":"https://openalex.org/S4210212798","display_name":"Foundations of Data Science","issn_l":"2639-8001","issn":["2639-8001"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-c","jufo-1","norway-1","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310315844","host_organization_name":"American Institute of Mathematical Sciences","host_organization_lineage":["https://openalex.org/P4310315844"],"host_organization_lineage_names":["American Institute of Mathematical Sciences"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Foundations of Data Science","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.3934/fods.2019019","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5002315948","display_name":"Benedict Leimkuhler","orcid":"https://orcid.org/0000-0001-8504-7091","observed_orcids":["https://orcid.org/0000-0001-8504-7091"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Benedict Leimkuhler","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Charles Matthews","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Charles Matthews","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5068323517","display_name":"Tiffany Vlaar","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tiffany Vlaar","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5002315948"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.0061,"has_fulltext":true,"cited_by_count":38,"citation_normalized_percentile":{"value":0.79664003,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":88,"max":100},"biblio":{"volume":"1","issue":"4","first_page":"457","last_page":"489"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.948199987411499,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.948199987411499,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12056","display_name":"Markov Chains and Monte Carlo Methods","score":0.03350000083446503,"subfield":{"id":"https://openalex.org/subfields/2613","display_name":"Statistics and Probability"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.007699999958276749,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/neural-network-training","display_name":"neural network training","score":0.828000009059906},{"id":"https://openalex.org/keywords/langevin-dynamics","display_name":"Langevin dynamics","score":0.7129999995231628},{"id":"https://openalex.org/keywords/parameter-space-sampling","display_name":"parameter space sampling","score":0.6830000281333923},{"id":"https://openalex.org/keywords/loss-landscape","display_name":"loss landscape","score":0.5009999871253967},{"id":"https://openalex.org/keywords/convergence-analysis","display_name":"convergence analysis","score":0.5519999861717224},{"id":"https://openalex.org/keywords/feedforward-neural-networks","display_name":"feedforward neural networks","score":0.6669999957084656}],"concepts":[{"id":"https://openalex.org/C51955184","wikidata":"https://www.wikidata.org/wiki/Q1545585","display_name":"Stochastic differential equation","level":2,"score":0.7276723980903625},{"id":"https://openalex.org/C73000952","wikidata":"https://www.wikidata.org/wiki/Q17007827","display_name":"Discretization","level":2,"score":0.6856284737586975},{"id":"https://openalex.org/C2780004032","wikidata":"https://www.wikidata.org/wiki/Q6485978","display_name":"Langevin dynamics","level":2,"score":0.6114209294319153},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6080259680747986},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5829538702964783},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.5487533211708069},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.5090422034263611},{"id":"https://openalex.org/C165464430","wikidata":"https://www.wikidata.org/wiki/Q1570441","display_name":"Parameterized complexity","level":2,"score":0.46901291608810425},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.45919644832611084},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4454857110977173},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.344757080078125},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3294144868850708},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.25329458713531494},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.23931097984313965},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.12031590938568115},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.08470022678375244},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0}],"mesh":[],"locations_count":6,"locations":[{"id":"doi:10.3934/fods.2019019","doi":"https://doi.org/10.3934/fods.2019019","is_oa":true,"landing_page_url":"https://doi.org/10.3934/fods.2019019","pdf_url":null,"source":{"id":"https://openalex.org/S4210212798","display_name":"Foundations of Data Science","issn_l":"2639-8001","issn":["2639-8001"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-c","jufo-1","norway-1","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310315844","host_organization_name":"American Institute of Mathematical Sciences","host_organization_lineage":["https://openalex.org/P4310315844"],"host_organization_lineage_names":["American Institute of Mathematical Sciences"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Foundations of Data Science","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:1908.11843","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/1908.11843","pdf_url":"https://arxiv.org/pdf/1908.11843","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"mag:2970936655","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1908.11843.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"pmh:oai:eprints.gla.ac.uk:310178","doi":null,"is_oa":true,"landing_page_url":"https://eprints.gla.ac.uk/310178/","pdf_url":null,"source":{"id":"https://openalex.org/S4210235606","display_name":"ENLIGHTEN (Jurnal Bimbingan dan Konseling Islam)","issn_l":"2622-8912","issn":["2622-8912","2622-8920"],"is_oa":true,"is_in_doaj":true,"is_core":false,"listed_in":["doaj"],"host_organization":"https://openalex.org/I4405259220","host_organization_name":"Institut Agama Islam Negeri Langsa","host_organization_lineage":["https://openalex.org/I4405259220"],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"Articles"},{"id":"pmh:oai:pure.ed.ac.uk:openaire/b202c373-6c5b-4340-be5f-a9b5f2250d96","doi":null,"is_oa":true,"landing_page_url":"https://www.research.ed.ac.uk/files/128890197/FoDS_0019_R1_LeMaVl2019_manuscript.pdf","pdf_url":null,"source":{"id":"https://openalex.org/S4306400321","display_name":"Edinburgh Research Explorer (University of Edinburgh)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I98677209","host_organization_name":"University of Edinburgh","host_organization_lineage":["https://openalex.org/I98677209"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Leimkuhler, B, Matthews, C & Vlaar, T 2019, 'Partitioned integrators for thermodynamic parameterization of neural networks', Foundations of Data Science, vol. 1, no. 4, pp. 457-489. https://doi.org/10.3934/fods.2019019","raw_type":"info:eu-repo/semantics/publishedVersion"},{"id":"doi:10.48550/arxiv.1908.11843","doi":"https://doi.org/10.48550/arxiv.1908.11843","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1908.11843","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.3934/fods.2019019","doi":"https://doi.org/10.3934/fods.2019019","is_oa":true,"landing_page_url":"https://doi.org/10.3934/fods.2019019","pdf_url":null,"source":{"id":"https://openalex.org/S4210212798","display_name":"Foundations of Data Science","issn_l":"2639-8001","issn":["2639-8001"],"is_oa":true,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","dongbi-c","jufo-1","norway-1","poland-20","russia-white-list-1"],"host_organization":"https://openalex.org/P4310315844","host_organization_name":"American Institute of Mathematical Sciences","host_organization_lineage":["https://openalex.org/P4310315844"],"host_organization_lineage_names":["American Institute of Mathematical Sciences"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Foundations of Data Science","raw_type":"journal-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1633672591","display_name":null,"funder_award_id":"N510129","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G2164220759","display_name":null,"funder_award_id":"EP/L016508","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G2500306150","display_name":null,"funder_award_id":"EP/L016508/01","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G4033230319","display_name":"Data-Driven Coarse-Graining using Space-Time Diffusion Maps","funder_award_id":"EP/P006175/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G444630160","display_name":null,"funder_award_id":"EP/N510129/","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G4500485203","display_name":"The Alan Turing Institute","funder_award_id":"EP/N510129/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"},{"id":"https://openalex.org/G8597009950","display_name":null,"funder_award_id":"EP/N510129/1","funder_id":"https://openalex.org/F4320334627","funder_display_name":"Engineering and Physical Sciences Research Council"}],"funders":[{"id":"https://openalex.org/F4320311518","display_name":"Heriot-Watt University","ror":"https://ror.org/04mghma93"},{"id":"https://openalex.org/F4320320064","display_name":"Scottish Funding Council","ror":"https://ror.org/056bwcz71"},{"id":"https://openalex.org/F4320334627","display_name":"Engineering and Physical Sciences Research Council","ror":"https://ror.org/0439y7842"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2970936655.pdf","grobid_xml":"https://content.openalex.org/works/W2970936655.grobid-xml"},"referenced_works_count":59,"referenced_works":["https://openalex.org/W6908809","https://openalex.org/W1499021337","https://openalex.org/W1503398984","https://openalex.org/W1565709818","https://openalex.org/W1567512734","https://openalex.org/W1642251103","https://openalex.org/W1677182931","https://openalex.org/W1850240193","https://openalex.org/W1899249567","https://openalex.org/W1921523184","https://openalex.org/W1954227087","https://openalex.org/W1979863594","https://openalex.org/W1983452151","https://openalex.org/W1983628095","https://openalex.org/W1987483432","https://openalex.org/W1991794210","https://openalex.org/W1998754480","https://openalex.org/W2011873184","https://openalex.org/W2017196167","https://openalex.org/W2024060531","https://openalex.org/W2047028564","https://openalex.org/W2048971218","https://openalex.org/W2063530260","https://openalex.org/W2079184994","https://openalex.org/W2104049095","https://openalex.org/W2115252128","https://openalex.org/W2123337239","https://openalex.org/W2127265926","https://openalex.org/W2128086871","https://openalex.org/W2135046866","https://openalex.org/W2146502635","https://openalex.org/W2146989110","https://openalex.org/W2165652770","https://openalex.org/W2167433878","https://openalex.org/W2258658489","https://openalex.org/W2527189750","https://openalex.org/W2546302380","https://openalex.org/W2554535786","https://openalex.org/W2564807118","https://openalex.org/W2593267444","https://openalex.org/W2604901683","https://openalex.org/W2626325961","https://openalex.org/W2768267830","https://openalex.org/W2769265955","https://openalex.org/W2774865485","https://openalex.org/W2891433050","https://openalex.org/W2899771611","https://openalex.org/W2902907165","https://openalex.org/W2939461465","https://openalex.org/W2948920178","https://openalex.org/W2950220847","https://openalex.org/W2963008903","https://openalex.org/W2963208657","https://openalex.org/W2963695615","https://openalex.org/W2964121744","https://openalex.org/W2969968290","https://openalex.org/W2971393380","https://openalex.org/W3100981869","https://openalex.org/W4235722944"],"related_works":["https://openalex.org/W6959136128","https://openalex.org/W4416159192","https://openalex.org/W7167795275","https://openalex.org/W7139145394","https://openalex.org/W2750949993","https://openalex.org/W4415761007","https://openalex.org/W4310124277","https://openalex.org/W2922737435","https://openalex.org/W2091154614","https://openalex.org/W4379924477"],"abstract_inverted_index":{"Traditionally,":[0],"neural":[1,68,112],"networks":[2],"are":[3,108],"parameterized":[4],"using":[5,141],"optimization":[6],"procedures":[7,17],"such":[8,156],"as":[9,41,157],"stochastic":[10,49,104,158],"gradient":[11,159],"descent,":[12],"RMSProp":[13],"and":[14,124,127,131,144,150,161,177],"ADAM.":[15,162],"These":[16],"tend":[18],"to":[19,39,110,153],"drive":[20],"the":[21,24,63,74,121,136],"parameters":[22,75],"of":[23,138],"network":[25,69],"toward":[26],"a":[27,53,115],"local":[28],"minimum.":[29],"In":[30],"this":[31],"article,":[32],"we":[33,81,134],"employ":[34],"alternative":[35],"'sampling'":[36],"algorithms":[37,183],"(referred":[38],"here":[40],"'thermodynamic":[42],"parameterization":[43,168],"methods')":[44],"which":[45,107],"rely":[46],"on":[47,57,77,102],"discretized":[48,103],"differential":[50,105],"equations":[51],"for":[52,88],"defined":[54],"target":[55],"distribution":[56],"parameter":[58],"space.":[59],"We":[60,163],"show":[61],"that":[62,166],"thermodynamic":[64,167],"perspective":[65],"already":[66],"improves":[67],"training.":[70],"Moreover,":[71],"by":[72],"partitioning":[73],"based":[76,101],"natural":[78],"layer":[79],"structure":[80],"obtain":[82],"schemes":[83],"with":[84,91],"very":[85],"rapid":[86],"convergence":[87,137],"data":[89],"sets":[90],"complicated":[92],"loss":[93],"landscapes.We":[94],"describe":[95],"easy-to-implement":[96],"hybrid":[97],"partitioned":[98],"numerical":[99,142],"algorithms,":[100],"equations,":[106],"adapted":[109],"feed-forward":[111],"networks,":[113],"including":[114],"multi-layer":[116],"Langevin":[117,123,125,130],"algorithm,":[118],"AdLaLa":[119],"(combining":[120,129],"adaptive":[122],"algorithms)":[126],"LOL":[128],"Overdamped":[132],"Langevin);":[133],"examine":[135],"these":[139],"methods":[140,169],"studies":[143],"compare":[145],"their":[146],"performance":[147],"among":[148],"themselves":[149],"in":[151],"relation":[152],"standard":[154,182],"alternatives":[155],"descent":[160],"present":[164],"evidence":[165],"can":[170],"be":[171],"(\u2170)":[172],"faster,":[173],"(\u2171)":[174],"more":[175,179],"accurate,":[176],"(\u2172)":[178],"robust":[180],"than":[181],"used":[184],"within":[185],"machine":[186],"learning":[187],"frameworks.":[188]},"counts_by_year":[{"year":2023,"cited_by_count":28},{"year":2022,"cited_by_count":1},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":5},{"year":2019,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2019-09-05T00:00:00"}