{"id":"https://openalex.org/W7201890201","doi":"https://doi.org/10.48550/arxiv.2608.05680","title":"Curvature Residual Geometry in Bregman Regression","display_name":"Curvature Residual Geometry in Bregman Regression","publication_year":2026,"publication_date":"2026-08-06","ids":{"openalex":"https://openalex.org/W7201890201","doi":"https://doi.org/10.48550/arxiv.2608.05680"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.05680","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05680","pdf_url":"https://arxiv.org/pdf/2608.05680","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/2608.05680","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5016120331","display_name":"Vu Khac Ky","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I109689652","display_name":"FPT University","ror":"https://ror.org/03esj4g97","country_code":"VN","type":"education","lineage":["https://openalex.org/I109689652"]}],"countries":["VN"],"is_corresponding":true,"raw_author_name":"Ky, Vu Khac","raw_affiliation_strings":["Mathematics Department , FPT University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Mathematics Department , FPT University","institution_ids":["https://openalex.org/I109689652"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5016120331"],"corresponding_institution_ids":["https://openalex.org/I109689652"],"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":0.7226999998092651,"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":0.7226999998092651,"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":0.07000000029802322,"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/T10136","display_name":"Statistical Methods and Inference","score":0.040300000458955765,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/curvature","display_name":"curvature","score":0.3149999976158142},{"id":"https://openalex.org/keywords/hessian-matrix","display_name":"Hessian matrix","score":0.5619999766349792},{"id":"https://openalex.org/keywords/gradient-descent-convergence","display_name":"gradient descent convergence","score":0.6830000281333923},{"id":"https://openalex.org/keywords/nonconvex-optimization","display_name":"non-convex optimization","score":0.7540000081062317},{"id":"https://openalex.org/keywords/step-size-conditions","display_name":"step size conditions","score":0.4099999964237213},{"id":"https://openalex.org/keywords/linear-convergence","display_name":"linear convergence","score":0.3529999852180481}],"concepts":[{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.9039000272750854},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.8263999819755554},{"id":"https://openalex.org/C72134830","wikidata":"https://www.wikidata.org/wiki/Q5166524","display_name":"Convexity","level":2,"score":0.6869000196456909},{"id":"https://openalex.org/C195065555","wikidata":"https://www.wikidata.org/wiki/Q214881","display_name":"Curvature","level":2,"score":0.5277000069618225},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.5113999843597412},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.47110000252723694},{"id":"https://openalex.org/C149073432","wikidata":"https://www.wikidata.org/wiki/Q4960382","display_name":"Bregman divergence","level":2,"score":0.43849998712539673},{"id":"https://openalex.org/C57691317","wikidata":"https://www.wikidata.org/wiki/Q1289248","display_name":"Scalar (mathematics)","level":2,"score":0.438400000333786},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.4147000014781952},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.4108999967575073},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.4034000039100647},{"id":"https://openalex.org/C73729460","wikidata":"https://www.wikidata.org/wiki/Q17023339","display_name":"Hessian equation","level":4,"score":0.39250001311302185},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.3919999897480011},{"id":"https://openalex.org/C130432447","wikidata":"https://www.wikidata.org/wiki/Q11420049","display_name":"Quartic function","level":2,"score":0.37549999356269836},{"id":"https://openalex.org/C48921125","wikidata":"https://www.wikidata.org/wiki/Q10861030","display_name":"Linear regression","level":2,"score":0.30140000581741333},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.29589998722076416},{"id":"https://openalex.org/C46889948","wikidata":"https://www.wikidata.org/wiki/Q2755024","display_name":"Nonlinear regression","level":3,"score":0.29510000348091125},{"id":"https://openalex.org/C57869625","wikidata":"https://www.wikidata.org/wiki/Q1783502","display_name":"Rate of convergence","level":3,"score":0.29499998688697815},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.29249998927116394},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C33257320","wikidata":"https://www.wikidata.org/wiki/Q1932744","display_name":"Second derivative","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C49712288","wikidata":"https://www.wikidata.org/wiki/Q77601250","display_name":"Positive-definite matrix","level":3,"score":0.27129998803138733},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.26750001311302185},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.26589998602867126},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.2515000104904175},{"id":"https://openalex.org/C2778067643","wikidata":"https://www.wikidata.org/wiki/Q166507","display_name":"Interval (graph theory)","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.05680","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05680","pdf_url":"https://arxiv.org/pdf/2608.05680","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.2608.05680","doi":"https://doi.org/10.48550/arxiv.2608.05680","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.05680","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:2608.05680","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05680","pdf_url":"https://arxiv.org/pdf/2608.05680","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":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7201890201.pdf","grobid_xml":"https://content.openalex.org/works/W7201890201.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7167727884","https://openalex.org/W2567697964","https://openalex.org/W4319303464","https://openalex.org/W2035507034","https://openalex.org/W7150788928","https://openalex.org/W2774022454","https://openalex.org/W4226101650","https://openalex.org/W4399115785","https://openalex.org/W7081575489","https://openalex.org/W4309208344"],"abstract_inverted_index":{"We":[0],"consider":[1],"linear":[2,31,105],"regression":[3,44],"models":[4],"fitted":[5,68],"by":[6],"minimizing":[7],"Bregman":[8,194],"losses":[9],"of":[10,58,66,86,107,124,163],"the":[11,24,30,35,42,56,59,63,67,84,87,90,111,122,147,154,161,173,185,189],"form":[12],"\\[":[13],"\\frac{1}{n}\\sum_{i=1}^n":[14],"\\left[":[15],"\u03d5(y_i)-\u03d5(x_i^\\top\u03b8)":[16],"-\u03d5'(x_i^\\top\u03b8)":[17],"\\bigl(y_i-x_i^\\top\u03b8\\bigr)":[18],"\\right],":[19],"\\]":[20],"where":[21],"\\(y_i\\)":[22],"is":[23,29,39],"observed":[25],"response":[26],"and":[27,62,89,103,127,130,188],"\\(x_i^\\top\u03b8\\)":[28],"prediction.":[32],"Even":[33],"when":[34],"generating":[36,60,186],"potential":[37,61,88,187],"\\(\u03d5\\)":[38],"strongly":[40],"convex,":[41],"resulting":[43],"objective":[45],"may":[46,144],"be":[47,73],"nonconvex":[48],"in":[49,192],"\\(\u03b8\\).":[50],"This":[51,93],"creates":[52],"a":[53,76],"gap":[54],"between":[55],"convexity":[57,119],"optimization":[64],"geometry":[65],"model.":[69],"The":[70],"Hessian":[71,149],"can":[72],"written":[74],"as":[75,172],"weighted":[77],"Gram":[78],"matrix":[79,191],"whose":[80],"weights":[81],"depend":[82],"on":[83],"derivatives":[85],"current":[91],"residuals.":[92],"representation":[94],"gives":[95],"simple":[96],"conditions":[97,133,143],"for":[98,134],"local":[99,129],"strong":[100],"convexity,":[101],"smoothness,":[102],"conditional":[104],"convergence":[106,170],"gradient":[108],"descent.":[109],"For":[110],"quadratic--quartic":[112],"potential,":[113],"we":[114],"derive":[115],"an":[116],"exact":[117],"scalar":[118,142],"condition,":[120],"identify":[121],"interval":[123],"negative":[125],"curvature,":[126],"obtain":[128],"global":[131],"sufficient":[132],"positive":[135,151],"curvature.":[136],"Numerical":[137],"experiments":[138],"indicate":[139,159],"that":[140,160],"these":[141],"fail":[145],"while":[146],"full":[148],"remains":[150],"definite":[152],"at":[153],"evaluated":[155],"points.":[156],"They":[157],"also":[158],"range":[162],"tested":[164],"gradient-descent":[165],"step":[166],"sizes":[167],"leading":[168],"to":[169],"decreases":[171],"quartic":[174],"parameter":[175],"grows.":[176],"These":[177],"results":[178],"characterize":[179],"how":[180],"residual-dependent":[181],"curvature":[182],"interacts":[183],"with":[184],"design":[190],"reverse":[193],"regression.":[195]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-08T00:00:00"}