{"id":"https://openalex.org/W4417070075","doi":"https://doi.org/10.48550/arxiv.2509.08483","title":"Modified Loss of Momentum Gradient Descent: Fine-Grained Analysis","display_name":"Modified Loss of Momentum Gradient Descent: Fine-Grained Analysis","publication_year":2025,"publication_date":"2025-09-10","ids":{"openalex":"https://openalex.org/W4417070075","doi":"https://doi.org/10.48550/arxiv.2509.08483"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2509.08483","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2509.08483","pdf_url":"https://arxiv.org/pdf/2509.08483","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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/2509.08483","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Cattaneo, Matias D.","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cattaneo, Matias D.","raw_affiliation_strings":["Princeton University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Princeton University","institution_ids":["https://openalex.org/I20089843"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5124872842","display_name":"Boris Shigida","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I20089843","display_name":"Princeton University","ror":"https://ror.org/00hx57361","country_code":"US","type":"education","lineage":["https://openalex.org/I20089843"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shigida, Boris","raw_affiliation_strings":["Princeton University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Princeton University","institution_ids":["https://openalex.org/I20089843"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I20089843"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":4,"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":0.9940999746322632,"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":0.9940999746322632,"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/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.0013000000035390258,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"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/T10545","display_name":"Optimization and Variational Analysis","score":0.0003000000142492354,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/heavy-ball-momentum","display_name":"heavy-ball momentum","score":0.9290000200271606},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"gradient descent","score":0.8069999814033508},{"id":"https://openalex.org/keywords/trajectory-approximation","display_name":"trajectory approximation","score":0.6060000061988831},{"id":"https://openalex.org/keywords/narayana-polynomials","display_name":"Narayana polynomials","score":0.8240000009536743},{"id":"https://openalex.org/keywords/eulerian-polynomials","display_name":"Eulerian polynomials","score":0.9860000014305115},{"id":"https://openalex.org/keywords/h-polynomials","display_name":"h-polynomials","score":0.8450000286102295},{"id":"https://openalex.org/keywords/graph-associahedra","display_name":"graph associahedra","score":0.9779999852180481},{"id":"https://openalex.org/keywords/mini-batch-optimization","display_name":"mini-batch optimization","score":0.8880000114440918}],"concepts":[{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.7421000003814697},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.6575999855995178},{"id":"https://openalex.org/C167879884","wikidata":"https://www.wikidata.org/wiki/Q727568","display_name":"Balanced flow","level":2,"score":0.6061000227928162},{"id":"https://openalex.org/C60718061","wikidata":"https://www.wikidata.org/wiki/Q1414747","display_name":"Momentum (technical analysis)","level":2,"score":0.5774000287055969},{"id":"https://openalex.org/C2780428219","wikidata":"https://www.wikidata.org/wiki/Q16952335","display_name":"Cover (algebra)","level":2,"score":0.5285999774932861},{"id":"https://openalex.org/C151376022","wikidata":"https://www.wikidata.org/wiki/Q168698","display_name":"Exponential function","level":2,"score":0.48590001463890076},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.4691999852657318},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.4458000063896179},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.44020000100135803},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.41929998993873596},{"id":"https://openalex.org/C182306322","wikidata":"https://www.wikidata.org/wiki/Q1779371","display_name":"Order (exchange)","level":2,"score":0.41350001096725464},{"id":"https://openalex.org/C43058520","wikidata":"https://www.wikidata.org/wiki/Q624580","display_name":"Eulerian path","level":3,"score":0.4097999930381775},{"id":"https://openalex.org/C75235859","wikidata":"https://www.wikidata.org/wiki/Q582659","display_name":"Exponential growth","level":2,"score":0.4047999978065491},{"id":"https://openalex.org/C529865628","wikidata":"https://www.wikidata.org/wiki/Q1790740","display_name":"Manifold (fluid mechanics)","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.3483999967575073},{"id":"https://openalex.org/C2776637919","wikidata":"https://www.wikidata.org/wiki/Q624380","display_name":"Descent (aeronautics)","level":2,"score":0.3441999852657318},{"id":"https://openalex.org/C180188523","wikidata":"https://www.wikidata.org/wiki/Q574576","display_name":"Exponential decay","level":2,"score":0.2782999873161316},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.2782000005245209},{"id":"https://openalex.org/C2777299769","wikidata":"https://www.wikidata.org/wiki/Q3707858","display_name":"Type (biology)","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.2639000117778778},{"id":"https://openalex.org/C2986878899","wikidata":"https://www.wikidata.org/wiki/Q11216","display_name":"Numerical approximation","level":3,"score":0.26339998841285706},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.2565000057220459},{"id":"https://openalex.org/C195906000","wikidata":"https://www.wikidata.org/wiki/Q1191722","display_name":"Matrix exponential","level":3,"score":0.25360000133514404}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2509.08483","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2509.08483","pdf_url":"https://arxiv.org/pdf/2509.08483","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2509.08483","doi":"https://doi.org/10.48550/arxiv.2509.08483","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2509.08483","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2509.08483","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2509.08483","pdf_url":"https://arxiv.org/pdf/2509.08483","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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/G78141540","display_name":"Partitioning-Based Learning Methods for Treatment Effect Estimation and Inference","funder_award_id":"2241575","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7826521039","display_name":"New Developments in Methodology for Program Evaluation","funder_award_id":"2019432","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320308774","display_name":"John Simon Guggenheim Memorial Foundation","ror":"https://ror.org/0407tnq23"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417070075.pdf","grobid_xml":"https://content.openalex.org/works/W4417070075.grobid-xml"},"referenced_works_count":3,"referenced_works":["https://openalex.org/W2799042347","https://openalex.org/W4390307242","https://openalex.org/W4405956499"],"related_works":["https://openalex.org/W4283453263","https://openalex.org/W3129088206","https://openalex.org/W4399455183","https://openalex.org/W4319165847","https://openalex.org/W3121837399","https://openalex.org/W2951958592","https://openalex.org/W2578906089","https://openalex.org/W4392209319","https://openalex.org/W4416719856","https://openalex.org/W1984402599"],"abstract_inverted_index":{"We":[0,88,126],"analyze":[1],"gradient":[2,42],"descent":[3,43],"with":[4,44],"Polyak":[5],"(1964)":[6],"heavy-ball":[7],"momentum":[8,12],"(HB)":[9],"whose":[10],"fixed":[11],"hyperparameter":[13],"$\u03b2\\in":[14],"(0,":[15],"1)$":[16],"provides":[17],"exponential":[18],"decay":[19],"of":[20,94,101,109,133,138,147,186,195],"memory.":[21],"Building":[22],"on":[23,31,182],"Kovachki":[24],"and":[25,78,117,123,155,174,188],"Stuart":[26],"(2021),":[27],"we":[28,67,142],"prove":[29,79,127],"that":[30,49,128,159],"an":[32],"exponentially":[33],"attractive":[34],"invariant":[35],"manifold":[36],"the":[37,50,58,95,98,139,156,161,183],"algorithm":[38],"is":[39,54],"exactly":[40],"plain":[41],"a":[45,64,91,106,190],"modified":[46,59,145],"loss,":[47],"provided":[48],"step":[51],"size":[52],"$h$":[53],"small":[55],"enough.":[56],"Although":[57],"loss":[60],"does":[61],"not":[62],"admit":[63],"closed-form":[65],"expression,":[66],"describe":[68],"it":[69],"up":[70],"to":[71],"$O(h^{\\mathcal{R}})$-errors":[72],"for":[73,192],"arbitrary":[74,148],"finite":[75,149],"order":[76,151],"$\\mathcal{R}$,":[77],"global":[80],"(finite":[81],"\"time\"":[82],"horizon)":[83],"trajectory":[84],"approximation":[85,150],"bounds":[86],"$O(h^{\\mathcal{R}})$.":[87],"then":[89],"conduct":[90],"fine-grained":[92],"analysis":[93,194],"combinatorics":[96],"underlying":[97],"memoryless":[99],"approximations":[100],"HB,":[102],"in":[103,111,120],"particular,":[104],"finding":[105],"rich":[107],"family":[108],"polynomials":[110,130],"$\u03b2$":[112],"hidden":[113],"inside":[114],"which":[115],"include":[116],"lie":[118],"coefficient-wise":[119],"between":[121],"Eulerian":[122],"Narayana":[124],"polynomials.":[125],"these":[129],"are":[131],"$h$-polynomials":[132],"certain":[134],"graph-associahedra.":[135],"As":[136],"corollaries":[137],"main":[140,184],"results,":[141],"derive":[143],"continuous":[144],"equations":[146],"(with":[152],"rigorous":[153],"bounds)":[154],"principal":[157],"flow":[158],"approximates":[160],"HB":[162,187],"dynamics,":[163],"generalizing":[164],"Rosca":[165],"et":[166],"al.":[167],"(2023).":[168],"Approximation":[169],"theorems":[170],"cover":[171],"both":[172],"full-batch":[173],"mini-batch":[175],"HB.":[176],"The":[177],"results":[178],"shed":[179],"new":[180],"light":[181],"features":[185],"outline":[189],"roadmap":[191],"similar":[193],"other":[196],"optimization":[197],"algorithms.":[198]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-10-10T00:00:00"}