{"id":"https://openalex.org/W7223017705","doi":"https://doi.org/10.48550/arxiv.2610.25335","title":"A Distribution-Free Separation Between Learned and Fixed Features","display_name":"A Distribution-Free Separation Between Learned and Fixed Features","publication_year":2026,"publication_date":"2026-10-18","ids":{"openalex":"https://openalex.org/W7223017705","doi":"https://doi.org/10.48550/arxiv.2610.25335"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.25335","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.25335","pdf_url":"https://arxiv.org/pdf/2610.25335","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/2610.25335","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5090634474","display_name":"Daniel Barzilai","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I53964585","display_name":"Weizmann Institute of Science","ror":"https://ror.org/0316ej306","country_code":"IL","type":"education","lineage":["https://openalex.org/I53964585"]}],"countries":["IL"],"is_corresponding":true,"raw_author_name":"Barzilai, Daniel","raw_affiliation_strings":["Weizmann Institute of Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Weizmann Institute of Science","institution_ids":["https://openalex.org/I53964585"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5090634474"],"corresponding_institution_ids":["https://openalex.org/I53964585"],"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/T10320","display_name":"Neural Networks and Applications","score":0.42239999771118164,"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/T10320","display_name":"Neural Networks and Applications","score":0.42239999771118164,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.38109999895095825,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.15950000286102295,"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/distribution-free-learning","display_name":"distribution-free learning","score":0.47999998927116394},{"id":"https://openalex.org/keywords/hypothesis-classes","display_name":"hypothesis classes","score":0.6190000176429749},{"id":"https://openalex.org/keywords/linear-predictor","display_name":"linear predictor","score":0.5400000214576721},{"id":"https://openalex.org/keywords/runtime-complexity","display_name":"runtime complexity","score":0.5759999752044678}],"concepts":[{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.7615000009536743},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6729000210762024},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5824999809265137},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5419999957084656},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5396000146865845},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5368000268936157},{"id":"https://openalex.org/C119322782","wikidata":"https://www.wikidata.org/wiki/Q2662236","display_name":"VC dimension","level":2,"score":0.5364999771118164},{"id":"https://openalex.org/C90119067","wikidata":"https://www.wikidata.org/wiki/Q43260","display_name":"Polynomial","level":2,"score":0.5238999724388123},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5065000057220459},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4853000044822693},{"id":"https://openalex.org/C311688","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Time complexity","level":2,"score":0.42829999327659607},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.4077000021934509},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.3856000006198883},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3659000098705292},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.35190001130104065},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3513000011444092},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3174000084400177},{"id":"https://openalex.org/C49766605","wikidata":"https://www.wikidata.org/wiki/Q207643","display_name":"Linear map","level":2,"score":0.2989000082015991},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.2743000090122223},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C136979486","wikidata":"https://www.wikidata.org/wiki/Q773483","display_name":"Existential quantification","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.25335","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.25335","pdf_url":"https://arxiv.org/pdf/2610.25335","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.2610.25335","doi":"https://doi.org/10.48550/arxiv.2610.25335","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.25335","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:2610.25335","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.25335","pdf_url":"https://arxiv.org/pdf/2610.25335","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":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7223017705.pdf","grobid_xml":"https://content.openalex.org/works/W7223017705.grobid-xml"},"referenced_works_count":2,"referenced_works":["https://openalex.org/W4414634274","https://openalex.org/W7213613070"],"related_works":["https://openalex.org/W4320086284","https://openalex.org/W4415200437","https://openalex.org/W3024317845","https://openalex.org/W3006363000","https://openalex.org/W2753713566","https://openalex.org/W2963992434","https://openalex.org/W3192896295","https://openalex.org/W2926417365","https://openalex.org/W4226426379","https://openalex.org/W4226016381"],"abstract_inverted_index":{"The":[0],"success":[1],"of":[2,26,44,65],"neural":[3,124],"networks":[4],"is":[5,153],"largely":[6],"attributed":[7],"to":[8,11,34,131],"their":[9],"ability":[10],"learn":[12],"useful":[13],"representations":[14],"from":[15],"training":[16],"data.":[17],"However,":[18],"known":[19],"results":[20],"that":[21,52,68,122],"prove":[22],"a":[23,63,74,79,119],"computational":[24],"advantage":[25,54],"learned":[27,72],"features":[28],"over":[29,90],"fixed":[30,92],"ones":[31],"are":[32],"restricted":[33],"specific":[35],"input":[36,129],"distributions.":[37],"We":[38,61],"settle":[39],"the":[40,58,115,123,132,143,147],"COLT":[41],"open":[42],"question":[43],"Feldman,":[45],"Kamath,":[46],"and":[47,106],"Srebro":[48],"(2026)":[49],"by":[50,73,87,150],"showing":[51],"this":[53],"can":[55,69],"persist":[56],"in":[57,137,155],"distribution-free":[59],"setting.":[60],"construct":[62],"family":[64],"hypothesis":[66,120],"classes":[67],"be":[70,85],"efficiently":[71],"natural":[75],"gradient-based":[76],"algorithm":[77],"on":[78,114],"two-layer":[80],"ReLU":[81],"network,":[82],"but":[83],"cannot":[84],"represented":[86],"linear":[88,151],"predictors":[89],"any":[91,102,107,128],"feature":[93],"map":[94],"with":[95],"polynomial":[96],"dimension":[97,144,148],"(and":[98],"runtime).":[99],"Specifically,":[100],"for":[101],"sufficiently":[103,108],"small":[104],"error":[105,136],"large":[109],"runtime":[110],"budget":[111],"R":[112],"(depending":[113],"error),":[116],"there":[117],"exists":[118],"class":[121],"network":[125],"learns":[126],"under":[127],"distribution":[130],"desired":[133],"expected":[134],"classification":[135],"time":[138],"at":[139],"most":[140],"R,":[141],"while":[142],"complexity":[145],"(governing":[146],"needed":[149],"predictors)":[152],"superpolynomial":[154],"R.":[156]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-21T00:00:00"}