{"id":"https://openalex.org/W7162649891","doi":"https://doi.org/10.48550/arxiv.2605.28612","title":"Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent","display_name":"Learning High-Dimensional Parity Functions with Product Networks using Gradient Descent","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W7162649891","doi":"https://doi.org/10.48550/arxiv.2605.28612"},"language":"en","primary_location":{"id":"pmh:oai:HAL:hal-05690807v1","is_oa":false,"landing_page_url":"https://hal.science/hal-05690807","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ICML 2026: Forty-Third International Conference on Machine Learning, Jul 2026, Seoul (Korea), South Korea","raw_type":"info:eu-repo/semantics/conferenceObject"},"type":"conference-paper","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.28612","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5027137158","display_name":"Guillaume Larue","orcid":"https://orcid.org/0000-0003-1989-4246"},"institutions":[{"id":"https://openalex.org/I4210116035","display_name":"Oros (France)","ror":"https://ror.org/024xe4a86","country_code":"FR","type":"company","lineage":["https://openalex.org/I4210116035"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Larue, Guillaume","raw_affiliation_strings":["Orange Research, Meylan, France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Orange Research, Meylan, France","institution_ids":["https://openalex.org/I4210116035"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002301712","display_name":"Louis-Adrien Dufr\u00e9ne","orcid":null},"institutions":[{"id":"https://openalex.org/I4210116035","display_name":"Oros (France)","ror":"https://ror.org/024xe4a86","country_code":"FR","type":"company","lineage":["https://openalex.org/I4210116035"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Dufr\u00e8ne, Louis-Adrien","raw_affiliation_strings":["Orange Research, Meylan, France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Orange Research, Meylan, France","institution_ids":["https://openalex.org/I4210116035"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062229278","display_name":"Quentin Lampin","orcid":"https://orcid.org/0000-0002-6996-9739"},"institutions":[{"id":"https://openalex.org/I4210116035","display_name":"Oros (France)","ror":"https://ror.org/024xe4a86","country_code":"FR","type":"company","lineage":["https://openalex.org/I4210116035"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Lampin, Quentin","raw_affiliation_strings":["Orange Research, Meylan, France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Orange Research, Meylan, France","institution_ids":["https://openalex.org/I4210116035"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067519923","display_name":"Hadi Ghauch","orcid":"https://orcid.org/0000-0002-9442-671X"},"institutions":[{"id":"https://openalex.org/I12356871","display_name":"T\u00e9l\u00e9com Paris","ror":"https://ror.org/01naq7912","country_code":"FR","type":"education","lineage":["https://openalex.org/I12356871","https://openalex.org/I205703379","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Ghauch, Hadi","raw_affiliation_strings":["T\u00e9l\u00e9com Paris, Institut Polytechnique de Paris, Palaiseau, France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"T\u00e9l\u00e9com Paris, Institut Polytechnique de Paris, Palaiseau, France","institution_ids":["https://openalex.org/I12356871"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5030665091","display_name":"Ghaya Rekaya","orcid":null},"institutions":[{"id":"https://openalex.org/I12356871","display_name":"T\u00e9l\u00e9com Paris","ror":"https://ror.org/01naq7912","country_code":"FR","type":"education","lineage":["https://openalex.org/I12356871","https://openalex.org/I205703379","https://openalex.org/I4210145102"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Rekaya, Ghaya","raw_affiliation_strings":["T\u00e9l\u00e9com Paris, Institut Polytechnique de Paris, Palaiseau, France"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"T\u00e9l\u00e9com Paris, Institut Polytechnique de Paris, Palaiseau, France","institution_ids":["https://openalex.org/I12356871"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.59208426,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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/T12072","display_name":"Machine Learning and Algorithms","score":0.15639999508857727,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.15639999508857727,"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/T10142","display_name":"Formal Methods in Verification","score":0.11949999630451202,"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"}},{"id":"https://openalex.org/T10720","display_name":"Complexity and Algorithms in Graphs","score":0.0908999964594841,"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/artificial-neural-network","display_name":"Artificial neural network","score":0.6340000033378601},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.602400004863739},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.4909999966621399},{"id":"https://openalex.org/keywords/empirical-risk-minimization","display_name":"Empirical risk minimization","score":0.487199991941452},{"id":"https://openalex.org/keywords/boolean-function","display_name":"Boolean function","score":0.4860000014305115},{"id":"https://openalex.org/keywords/parity","display_name":"Parity (physics)","score":0.4643999934196472},{"id":"https://openalex.org/keywords/gradient-descent","display_name":"Gradient descent","score":0.4519999921321869},{"id":"https://openalex.org/keywords/polynomial","display_name":"Polynomial","score":0.3968999981880188},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.34850001335144043}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6340000033378601},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.602400004863739},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5263000130653381},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.4909999966621399},{"id":"https://openalex.org/C107321475","wikidata":"https://www.wikidata.org/wiki/Q5374254","display_name":"Empirical risk minimization","level":2,"score":0.487199991941452},{"id":"https://openalex.org/C187455244","wikidata":"https://www.wikidata.org/wiki/Q942353","display_name":"Boolean function","level":2,"score":0.4860000014305115},{"id":"https://openalex.org/C2777151079","wikidata":"https://www.wikidata.org/wiki/Q141160","display_name":"Parity (physics)","level":2,"score":0.4643999934196472},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.4519999921321869},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44620001316070557},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4104999899864197},{"id":"https://openalex.org/C90119067","wikidata":"https://www.wikidata.org/wiki/Q43260","display_name":"Polynomial","level":2,"score":0.3968999981880188},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.384799987077713},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3578000068664551},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.34850001335144043},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.34389999508857727},{"id":"https://openalex.org/C176248197","wikidata":"https://www.wikidata.org/wiki/Q458526","display_name":"Probably approximately correct learning","level":4,"score":0.34049999713897705},{"id":"https://openalex.org/C151376022","wikidata":"https://www.wikidata.org/wiki/Q168698","display_name":"Exponential function","level":2,"score":0.33799999952316284},{"id":"https://openalex.org/C108710211","wikidata":"https://www.wikidata.org/wiki/Q11538","display_name":"Mathematical proof","level":2,"score":0.3294999897480011},{"id":"https://openalex.org/C2778445095","wikidata":"https://www.wikidata.org/wiki/Q18354077","display_name":"Sample complexity","level":2,"score":0.32820001244544983},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.3215000033378601},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3183000087738037},{"id":"https://openalex.org/C206688291","wikidata":"https://www.wikidata.org/wiki/Q7617819","display_name":"Stochastic gradient descent","level":3,"score":0.3138999938964844},{"id":"https://openalex.org/C50292564","wikidata":"https://www.wikidata.org/wiki/Q2462783","display_name":"Computational learning theory","level":3,"score":0.30239999294281006},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.29260000586509705},{"id":"https://openalex.org/C3309909","wikidata":"https://www.wikidata.org/wiki/Q864155","display_name":"Binary decision diagram","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C55974624","wikidata":"https://www.wikidata.org/wiki/Q1188504","display_name":"Exponential family","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C10628310","wikidata":"https://www.wikidata.org/wiki/Q619458","display_name":"Orthogonal polynomials","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C40969351","wikidata":"https://www.wikidata.org/wiki/Q3516228","display_name":"Word error rate","level":2,"score":0.2529999911785126},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:HAL:hal-05690807v1","is_oa":false,"landing_page_url":"https://hal.science/hal-05690807","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL (Le Centre pour la Communication Scientifique Directe)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I1294671590","host_organization_name":"Centre National de la Recherche Scientifique","host_organization_lineage":["https://openalex.org/I1294671590"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"ICML 2026: Forty-Third International Conference on Machine Learning, Jul 2026, Seoul (Korea), South Korea","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.2605.28612","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28612","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.28612","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28612","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Parity":[0],"functions":[1,20],"are":[2],"fundamental":[3,112],"Boolean":[4],"operations":[5],"with":[6,55,61,73,89],"critical":[7],"applications":[8],"across":[9,82],"machine":[10,133],"learning,":[11,72],"cryptography,":[12],"and":[13,65,98,118,132],"error":[14],"correction.":[15],"Yet,":[16],"learning":[17,99,134],"high-dimensional":[18],"parity":[19,71],"poses":[21],"significant":[22],"challenges:":[23],"in":[24],"a":[25],"general":[26],"setting,":[27],"standard":[28],"neural":[29,52,125,130],"network":[30],"architectures":[31,53],"typically":[32],"require":[33],"exponential":[34],"sample":[35],"complexity,":[36],"making":[37],"gradient-based":[38],"optimization":[39],"intractable":[40],"for":[41,96,124],"large":[42],"number":[43],"of":[44,76],"inputs":[45,60],"$N$.":[46],"We":[47],"demonstrate":[48],"that":[49],"compact":[50],"product-based":[51],"combined":[54],"stochastic":[56],"data":[57,119],"sparsity":[58],"(Bernoulli":[59],"$p_e":[62],"\\leq":[63],"1/N$)":[64],"appropriate":[66],"hyperparameter":[67,94],"choice":[68],"enable":[69],"efficient":[70],"theoretical":[74],"guarantees":[75],"convergence.":[77],"Experiments":[78],"validate":[79],"our":[80],"theory":[81],"dimensions":[83],"up":[84],"to":[85,136],"$N":[86],"=":[87],"100{,}000$,":[88],"empirical":[90],"evidence":[91],"showing":[92],"optimal":[93],"choices":[95],"$p_e$":[97],"rate":[100],"$\u03b1$,":[101],"as":[102,104],"well":[103],"polynomial":[105],"complexity":[106],"scaling":[107],"laws.":[108],"This":[109],"work":[110],"establishes":[111],"connections":[113],"between":[114],"architectural":[115],"inductive":[116],"bias":[117],"sparsity,":[120],"opening":[121],"new":[122],"possibilities":[123],"arithmetic,":[126],"structured":[127],"reasoning,":[128],"binary":[129],"networks,":[131],"applied":[135],"automated":[137],"protocol":[138],"discovery.":[139]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-05-29T00:00:00"}