{"id":"https://openalex.org/W7212010029","doi":"https://doi.org/10.48550/arxiv.2609.08219","title":"Speed limits for information acquisition in stochastic learning dynamics","display_name":"Speed limits for information acquisition in stochastic learning dynamics","publication_year":2026,"publication_date":"2026-09-08","ids":{"openalex":"https://openalex.org/W7212010029","doi":"https://doi.org/10.48550/arxiv.2609.08219"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.08219","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.08219","pdf_url":"https://arxiv.org/pdf/2609.08219","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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.08219","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5150953212","display_name":"Shuta Kobayashi","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I22299242","display_name":"Kyoto University","ror":"https://ror.org/02kpeqv85","country_code":"JP","type":"education","lineage":["https://openalex.org/I22299242"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kobayashi, Shuta","raw_affiliation_strings":["Department of Physics , Kyoto University , Kyoto 606-8502 , Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Physics , Kyoto University , Kyoto 606-8502 , Japan","institution_ids":["https://openalex.org/I22299242"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5042876485","display_name":"Andreas Dechant","orcid":"https://orcid.org/0000-0002-5782-5893","observed_orcids":["https://orcid.org/0000-0002-5782-5893"]},"institutions":[{"id":"https://openalex.org/I22299242","display_name":"Kyoto University","ror":"https://ror.org/02kpeqv85","country_code":"JP","type":"education","lineage":["https://openalex.org/I22299242"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Dechant, Andreas","raw_affiliation_strings":["Department of Physics , Kyoto University , Kyoto 606-8502 , Japan"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Physics , Kyoto University , Kyoto 606-8502 , Japan","institution_ids":["https://openalex.org/I22299242"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I22299242"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.8125,"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":0.8125,"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/T10320","display_name":"Neural Networks and Applications","score":0.12370000034570694,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.02160000056028366,"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/fisher-information","display_name":"Fisher information","score":0.7099999785423279},{"id":"https://openalex.org/keywords/speed-limits","display_name":"speed limits","score":0.8730000257492065},{"id":"https://openalex.org/keywords/stochastic-gradient-descent","display_name":"stochastic gradient descent","score":0.9350000023841858},{"id":"https://openalex.org/keywords/information-acquisition","display_name":"information acquisition","score":0.7519999742507935},{"id":"https://openalex.org/keywords/latent-variables","display_name":"latent variables","score":0.7480000257492065},{"id":"https://openalex.org/keywords/information-theoretic-limits","display_name":"information-theoretic limits","score":0.6259999871253967},{"id":"https://openalex.org/keywords/learning-dynamics","display_name":"learning dynamics","score":0.5260000228881836}],"concepts":[{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.6132000088691711},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.5534999966621399},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5307000279426575},{"id":"https://openalex.org/C8272713","wikidata":"https://www.wikidata.org/wiki/Q176737","display_name":"Stochastic process","level":2,"score":0.48669999837875366},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47940000891685486},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.47609999775886536},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.4745999872684479},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.42489999532699585},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.41609999537467957},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4140999913215637},{"id":"https://openalex.org/C55479107","wikidata":"https://www.wikidata.org/wiki/Q97663916","display_name":"Stochastic approximation","level":3,"score":0.382099986076355},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.3804999887943268},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.35920000076293945},{"id":"https://openalex.org/C86582703","wikidata":"https://www.wikidata.org/wiki/Q7617824","display_name":"Stochastic neural network","level":4,"score":0.35679998993873596},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.34619998931884766},{"id":"https://openalex.org/C2779136372","wikidata":"https://www.wikidata.org/wiki/Q10283002","display_name":"Information flow","level":2,"score":0.3346000015735626},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.3212999999523163},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.31630000472068787},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3086000084877014},{"id":"https://openalex.org/C145912823","wikidata":"https://www.wikidata.org/wiki/Q113558","display_name":"Dynamics (music)","level":2,"score":0.2946999967098236},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.2892000079154968},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2825999855995178},{"id":"https://openalex.org/C127491075","wikidata":"https://www.wikidata.org/wiki/Q7617825","display_name":"Stochastic modelling","level":2,"score":0.2685999870300293},{"id":"https://openalex.org/C158535547","wikidata":"https://www.wikidata.org/wiki/Q5165437","display_name":"Continuous-time stochastic process","level":3,"score":0.26109999418258667},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.2533000111579895}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.08219","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.08219","pdf_url":"https://arxiv.org/pdf/2609.08219","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":"doi:10.48550/arxiv.2609.08219","doi":"https://doi.org/10.48550/arxiv.2609.08219","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.08219","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":"pmh:oai:arXiv.org:2609.08219","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.08219","pdf_url":"https://arxiv.org/pdf/2609.08219","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"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G3703446686","display_name":null,"funder_award_id":"24H00833","funder_id":"https://openalex.org/F4320334764","funder_display_name":"Japan Society for the Promotion of Science"},{"id":"https://openalex.org/G8575185436","display_name":"Fundamental constraints on response in classical and quantum out-of-equilibrium systems","funder_award_id":"25K00926","funder_id":"https://openalex.org/F4320334764","funder_display_name":"Japan Society for the Promotion of Science"}],"funders":[{"id":"https://openalex.org/F4320320879","display_name":"Deutsche Forschungsgemeinschaft","ror":"https://ror.org/018mejw64"},{"id":"https://openalex.org/F4320334764","display_name":"Japan Society for the Promotion of Science","ror":"https://ror.org/00hhkn466"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7212010029.pdf","grobid_xml":"https://content.openalex.org/works/W7212010029.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4394659030","https://openalex.org/W2157969264","https://openalex.org/W3126682903","https://openalex.org/W2970638103","https://openalex.org/W4417041451","https://openalex.org/W2914300050","https://openalex.org/W2765393161","https://openalex.org/W4385373972","https://openalex.org/W7104183320","https://openalex.org/W7103748654"],"abstract_inverted_index":{"Neural":[0],"networks":[1],"acquire":[2,37],"internal":[3],"representations":[4],"through":[5],"learning.":[6],"In":[7],"this":[8],"work,":[9],"we":[10],"formulate":[11],"stochastic":[12,19],"gradient":[13],"descent":[14],"(SGD)":[15],"as":[16,115],"a":[17,23,116],"Markovian":[18],"process":[20,126],"and":[21,55,66,94,132],"derive":[22],"Fisher-information":[24,112],"flow":[25,52],"speed":[26,113],"limit":[27],"that":[28],"bounds":[29],"the":[30,46,50,60,75,84,89,92,106,124,140],"rate":[31],"at":[32],"which":[33,99,121],"trainable":[34,141],"parameters":[35,102],"can":[36,136],"information":[38,51,85],"about":[39],"data-generating":[40,101,125],"parameters.":[41,108,142],"The":[42],"resulting":[43],"inequality":[44],"decomposes":[45],"upper":[47],"bound":[48,76,90],"on":[49],"into":[53],"drift":[54],"noise":[56],"contributions,":[57],"thereby":[58],"quantifying":[59],"roles":[61],"of":[62,123],"deterministic":[63],"learning":[64],"forces":[65],"SGD-induced":[67],"fluctuations":[68],"from":[69],"an":[70],"information-theoretic":[71],"perspective.":[72],"We":[73],"verify":[74],"in":[77,105,129,139],"analytically":[78],"tractable":[79],"basis-function":[80],"linear":[81],"regression,":[82],"where":[83],"budget":[86],"predicted":[87],"by":[88],"reproduces":[91],"ordering":[93],"characteristic":[95],"time":[96],"scales":[97],"with":[98],"different":[100],"are":[103,127],"encoded":[104,138],"learned":[107],"These":[109],"results":[110],"establish":[111],"limits":[114],"quantitative":[117],"framework":[118],"for":[119],"understanding":[120],"properties":[122],"acquired,":[128],"what":[130],"order,":[131],"how":[133],"rapidly":[134],"they":[135],"be":[137]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-10T00:00:00"}