{"id":"https://openalex.org/W7220916747","doi":"https://doi.org/10.48550/arxiv.2610.09196","title":"Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines","display_name":"Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines","publication_year":2026,"publication_date":"2026-10-06","ids":{"openalex":"https://openalex.org/W7220916747","doi":"https://doi.org/10.48550/arxiv.2610.09196"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.09196","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.09196","pdf_url":"https://arxiv.org/pdf/2610.09196","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2610.09196","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147224682","display_name":"Andrew Cheng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Cheng, Andrew","raw_affiliation_strings":["Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University","institution_ids":["https://openalex.org/I136199984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079855181","display_name":"Bobak T. Kiani","orcid":"https://orcid.org/0000-0003-1477-0308","observed_orcids":["https://orcid.org/0000-0003-1477-0308"]},"institutions":[{"id":"https://openalex.org/I135474949","display_name":"Bowdoin College","ror":"https://ror.org/03gh96r95","country_code":"US","type":"education","lineage":["https://openalex.org/I135474949"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kiani, Bobak T.","raw_affiliation_strings":["Bowdoin College"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Bowdoin College","institution_ids":["https://openalex.org/I135474949"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102945856","display_name":"Yue Lu","orcid":"https://orcid.org/0000-0003-2216-1035","observed_orcids":["https://orcid.org/0000-0003-2216-1035"]},"institutions":[{"id":"https://openalex.org/I136199984","display_name":"Harvard University","ror":"https://ror.org/03vek6s52","country_code":"US","type":"education","lineage":["https://openalex.org/I136199984"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Lu, Yue M.","raw_affiliation_strings":["Harvard University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harvard University","institution_ids":["https://openalex.org/I136199984"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051451341","display_name":"Adityanarayanan Radhakrishnan","orcid":"https://orcid.org/0000-0002-0878-815X","observed_orcids":["https://orcid.org/0000-0002-0878-815X"]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Radhakrishnan, Adityanarayanan","raw_affiliation_strings":["Massachusetts Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology","institution_ids":["https://openalex.org/I63966007"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"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":0.8461999893188477,"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":0.8461999893188477,"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/T10320","display_name":"Neural Networks and Applications","score":0.06239999830722809,"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":0.017500000074505806,"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/representation-learning","display_name":"representation learning","score":0.593999981880188},{"id":"https://openalex.org/keywords/exact-dynamics","display_name":"exact dynamics","score":0.6150000095367432},{"id":"https://openalex.org/keywords/finite-sample-analysis","display_name":"finite-sample analysis","score":0.7480000257492065},{"id":"https://openalex.org/keywords/trajectory-reconstruction","display_name":"trajectory reconstruction","score":0.7120000123977661},{"id":"https://openalex.org/keywords/multi-output-regression","display_name":"multi-output regression","score":0.7480000257492065},{"id":"https://openalex.org/keywords/tikhonov-regularization","display_name":"Tikhonov regularization","score":0.7429999709129333},{"id":"https://openalex.org/keywords/iteratively-reweighted-least-squares","display_name":"iteratively reweighted least squares","score":0.8399999737739563}],"concepts":[{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.7235999703407288},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.5446000099182129},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.5196999907493591},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5166000127792358},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.5130000114440918},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5041999816894531},{"id":"https://openalex.org/C48921125","wikidata":"https://www.wikidata.org/wiki/Q10861030","display_name":"Linear regression","level":2,"score":0.4876999855041504},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4731999933719635},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.46219998598098755},{"id":"https://openalex.org/C180623205","wikidata":"https://www.wikidata.org/wiki/Q1268589","display_name":"Outer product","level":3,"score":0.4503999948501587},{"id":"https://openalex.org/C203233044","wikidata":"https://www.wikidata.org/wiki/Q5264358","display_name":"Design matrix","level":3,"score":0.4465000033378601},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.436599999666214},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4187999963760376},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.41359999775886536},{"id":"https://openalex.org/C2780985081","wikidata":"https://www.wikidata.org/wiki/Q676502","display_name":"Data Matrix","level":5,"score":0.36959999799728394},{"id":"https://openalex.org/C83546350","wikidata":"https://www.wikidata.org/wiki/Q1139051","display_name":"Regression","level":2,"score":0.36559998989105225},{"id":"https://openalex.org/C2776639384","wikidata":"https://www.wikidata.org/wiki/Q840396","display_name":"Ideal (ethics)","level":2,"score":0.35929998755455017},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.3425000011920929},{"id":"https://openalex.org/C70518039","wikidata":"https://www.wikidata.org/wiki/Q16000077","display_name":"Dimensionality reduction","level":2,"score":0.33629998564720154},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3343999981880188},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.3127000033855438},{"id":"https://openalex.org/C32224588","wikidata":"https://www.wikidata.org/wiki/Q7250175","display_name":"Proper linear model","level":4,"score":0.30300000309944153},{"id":"https://openalex.org/C126090379","wikidata":"https://www.wikidata.org/wiki/Q6094424","display_name":"Iteratively reweighted least squares","level":4,"score":0.30300000309944153},{"id":"https://openalex.org/C152877465","wikidata":"https://www.wikidata.org/wiki/Q208042","display_name":"Regression analysis","level":2,"score":0.3003000020980835},{"id":"https://openalex.org/C9936470","wikidata":"https://www.wikidata.org/wiki/Q6510405","display_name":"Least-squares function approximation","level":3,"score":0.2985000014305115},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C131109320","wikidata":"https://www.wikidata.org/wiki/Q581012","display_name":"Linear prediction","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C149769383","wikidata":"https://www.wikidata.org/wiki/Q7520804","display_name":"Simple linear regression","level":3,"score":0.27889999747276306},{"id":"https://openalex.org/C49766605","wikidata":"https://www.wikidata.org/wiki/Q207643","display_name":"Linear map","level":2,"score":0.2782000005245209},{"id":"https://openalex.org/C21080849","wikidata":"https://www.wikidata.org/wiki/Q13611879","display_name":"Data point","level":2,"score":0.27480000257492065},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.271699994802475},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.27140000462532043},{"id":"https://openalex.org/C60316415","wikidata":"https://www.wikidata.org/wiki/Q6664520","display_name":"Local regression","level":4,"score":0.26499998569488525},{"id":"https://openalex.org/C37616216","wikidata":"https://www.wikidata.org/wiki/Q3218363","display_name":"Lasso (programming language)","level":2,"score":0.2639000117778778},{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C21556879","wikidata":"https://www.wikidata.org/wiki/Q43219517","display_name":"Moore\u2013Penrose pseudoinverse","level":3,"score":0.26030001044273376},{"id":"https://openalex.org/C6802819","wikidata":"https://www.wikidata.org/wiki/Q1072174","display_name":"Linear system","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C69738355","wikidata":"https://www.wikidata.org/wiki/Q1228929","display_name":"Linear discriminant analysis","level":2,"score":0.25369998812675476}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.09196","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.09196","pdf_url":"https://arxiv.org/pdf/2610.09196","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.2610.09196","doi":"https://doi.org/10.48550/arxiv.2610.09196","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.09196","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:2610.09196","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.09196","pdf_url":"https://arxiv.org/pdf/2610.09196","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/W7220916747.pdf","grobid_xml":"https://content.openalex.org/works/W7220916747.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4408928873","https://openalex.org/W7093576455","https://openalex.org/W4415796071","https://openalex.org/W4415879047","https://openalex.org/W3080149057","https://openalex.org/W3098901286","https://openalex.org/W7020090846","https://openalex.org/W4307001334","https://openalex.org/W4361230897","https://openalex.org/W3007945084"],"abstract_inverted_index":{"Recursive":[0],"feature":[1,34,111,133],"machines":[2],"(RFMs)":[3],"learn":[4],"representations":[5,49],"of":[6,20,60,80],"data":[7,71,148],"by":[8,75,153],"alternating":[9],"between":[10,31,88],"fitting":[11],"a":[12,15,43,76],"predictor":[13,22],"to":[14,100,115],"dataset":[16],"and":[17,33,58,72,91,145],"updating":[18],"features":[19,151],"that":[21,108],"using":[23],"the":[24,56,85,97,109,129,132,150],"average":[25],"gradient":[26],"outer":[27],"product":[28],"(AGOP).":[29],"Connections":[30],"AGOPs":[32],"learning":[35],"in":[36,63,131],"neural":[37],"networks":[38],"motivate":[39],"linear":[40,61,89,156],"RFMs":[41],"as":[42,136],"simple":[44,155],"setting":[45,99],"for":[46,124],"analyzing":[47],"how":[48],"evolve":[50],"during":[51],"training.":[52],"Here,":[53],"we":[54,127],"study":[55],"dynamics":[57],"statistics":[59],"RFM":[62,90],"noisy":[64],"multi-output":[65,102],"regression":[66,103],"with":[67,104,138],"isotropic":[68],"sub-Gaussian":[69],"input":[70],"targets":[73],"generated":[74],"low-rank":[77],"teacher":[78],"matrix":[79,112,134],"dimension":[81],"$d$.":[82],"We":[83,106],"extend":[84],"known":[86],"connection":[87],"iteratively":[92],"reweighted":[93],"least":[94],"squares":[95],"from":[96],"interpolating":[98],"ridge-regularized":[101],"noise.":[105],"show":[107,128],"learned":[110,152],"remains":[113],"close":[114],"its":[116],"infinite-data":[117],"ideal":[118],"counterpart":[119],"at":[120],"every":[121],"iteration.":[122],"Namely,":[123],"$n$":[125],"samples,":[126],"error":[130],"decays":[135],"$O(\\sqrt{d/n})$":[137],"high":[139],"probability.":[140],"Experiments":[141],"on":[142],"real-world":[143],"text":[144],"single-cell":[146],"gene-expression":[147],"illustrate":[149],"this":[154],"model.":[157]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-10-09T00:00:00"}