{"id":"https://openalex.org/W7168484903","doi":"https://doi.org/10.48550/arxiv.2607.13631","title":"How the Hessian-Spectrum of Neural Networks Depends on Data","display_name":"How the Hessian-Spectrum of Neural Networks Depends on Data","publication_year":2026,"publication_date":"2026-07-15","ids":{"openalex":"https://openalex.org/W7168484903","doi":"https://doi.org/10.48550/arxiv.2607.13631"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.13631","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13631","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2607.13631","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5108842148","display_name":"Jasraj Singh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Singh, Jasraj","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081765056","display_name":"Enea Monzio Compagnoni","orcid":"https://orcid.org/0009-0004-7094-2586"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Compagnoni, Enea Monzio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5140982780","display_name":"Antonio Orvieto","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Orvieto, Antonio","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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":0.9132999777793884,"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":0.9132999777793884,"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/T12676","display_name":"Machine Learning and ELM","score":0.01269999984651804,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.009499999694526196,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/hessian-matrix","display_name":"Hessian matrix","score":0.8914999961853027},{"id":"https://openalex.org/keywords/eigenvalues-and-eigenvectors","display_name":"Eigenvalues and eigenvectors","score":0.5813000202178955},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5593000054359436},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.41200000047683716},{"id":"https://openalex.org/keywords/deep-neural-networks","display_name":"Deep neural networks","score":0.3409999907016754},{"id":"https://openalex.org/keywords/interpretation","display_name":"Interpretation (philosophy)","score":0.28700000047683716},{"id":"https://openalex.org/keywords/empirical-research","display_name":"Empirical research","score":0.28290000557899475}],"concepts":[{"id":"https://openalex.org/C203616005","wikidata":"https://www.wikidata.org/wiki/Q620495","display_name":"Hessian matrix","level":2,"score":0.8914999961853027},{"id":"https://openalex.org/C158693339","wikidata":"https://www.wikidata.org/wiki/Q190524","display_name":"Eigenvalues and eigenvectors","level":2,"score":0.5813000202178955},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5644999742507935},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5593000054359436},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5479000210762024},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.41200000047683716},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.39969998598098755},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39480000734329224},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.37310001254081726},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3571000099182129},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.3409999907016754},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.32190001010894775},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.28700000047683716},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C184389593","wikidata":"https://www.wikidata.org/wiki/Q603159","display_name":"Curve fitting","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2669999897480011},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C145912823","wikidata":"https://www.wikidata.org/wiki/Q113558","display_name":"Dynamics (music)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.25380000472068787},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.13631","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13631","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2607.13631","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.13631","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,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0],"Hessian":[1,59],"matrix":[2],"is":[3,94],"an":[4,71],"important":[5],"quantity":[6],"of":[7,28,57,60,74,91,101,117],"interest":[8],"when":[9],"it":[10],"comes":[11],"to":[12,97,104,144,148],"studying":[13],"the":[14,55,58,89,92,98,115,119],"loss":[15],"landscape":[16],"and":[17,66,68,77,112],"optimization":[18],"dynamics":[19],"in":[20,139],"deep":[21],"learning,":[22],"as":[23,25,126,128],"well":[24,127],"designing":[26],"measures":[27],"generalization,":[29],"second-order":[30],"learning":[31,151],"algorithms,":[32],"etc.":[33],"Prior":[34],"works":[35],"have":[36],"focused":[37],"on":[38],"empirical":[39],"results":[40],"or":[41],"pursued":[42],"a":[43,124],"theoretical":[44],"treatment":[45],"under":[46],"overly":[47],"simplified":[48],"settings.":[49],"In":[50],"this":[51],"work,":[52],"we":[53,86],"derive":[54],"eigenvalues":[56],"linear":[61],"networks":[62],"with":[63,70,83],"arbitrary":[64,72],"widths":[65],"depths,":[67],"datasets":[69],"number":[73],"samples,":[75],"features,":[76],"labels.":[78],"Importantly,":[79],"for":[80],"classification":[81],"tasks":[82],"MSE":[84],"loss,":[85],"identify":[87],"that":[88,133],"sharpness":[90],"solution":[93],"directly":[95],"related":[96],"maximum":[99],"proportion":[100],"samples":[102],"belonging":[103],"any":[105],"class.":[106],"We":[107,131],"empirically":[108],"validate":[109],"our":[110,134,146],"predictions":[111,135],"systematically":[113],"analyze":[114],"effects":[116],"shedding":[118],"impractical":[120],"assumptions":[121],"one":[122],"at":[123],"time,":[125],"incorporating":[129],"nonlinearities.":[130],"observe":[132],"are":[136],"considerably":[137],"robust":[138],"most":[140],"cases,":[141],"allowing":[142],"us":[143],"extend":[145],"conclusions":[147],"more":[149],"practical":[150],"setups.":[152]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-17T00:00:00"}