{"id":"https://openalex.org/W7201870942","doi":"https://doi.org/10.48550/arxiv.2608.05464","title":"Effective pruning of task-trained recurrent neural networks using noisy fluctuations and connection rescaling","display_name":"Effective pruning of task-trained recurrent neural networks using noisy fluctuations and connection rescaling","publication_year":2026,"publication_date":"2026-08-05","ids":{"openalex":"https://openalex.org/W7201870942","doi":"https://doi.org/10.48550/arxiv.2608.05464"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.05464","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05464","pdf_url":"https://arxiv.org/pdf/2608.05464","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/2608.05464","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5145952076","display_name":"Sanjith Senthil","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Senthil, Sanjith","raw_affiliation_strings":["The Harker School , San Jose , CA 95129 , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Harker School , San Jose , CA 95129 , USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5011931383","display_name":"Rishidev Chaudhuri","orcid":"https://orcid.org/0000-0003-2301-3261","observed_orcids":["https://orcid.org/0000-0003-2301-3261"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chaudhuri, Rishidev","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":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":0.9419000148773193,"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":0.9419000148773193,"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/T10581","display_name":"Neural dynamics and brain function","score":0.01899999938905239,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.012600000016391277,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/neural-network-pruning","display_name":"neural network pruning","score":0.7839999794960022},{"id":"https://openalex.org/keywords/recurrent-neural-networks","display_name":"recurrent neural networks","score":0.9440000057220459},{"id":"https://openalex.org/keywords/connection-strength","display_name":"connection strength","score":0.5490000247955322},{"id":"https://openalex.org/keywords/activity-dependent-plasticity","display_name":"activity-dependent plasticity","score":0.5529999732971191}],"concepts":[{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.8671000003814697},{"id":"https://openalex.org/C147168706","wikidata":"https://www.wikidata.org/wiki/Q1457734","display_name":"Recurrent neural network","level":3,"score":0.7038000226020813},{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.6291999816894531},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.529699981212616},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.504800021648407},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.48399999737739563},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4684000015258789},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.4675999879837036},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4611999988555908},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4226999878883362},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.41929998993873596},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3578999936580658},{"id":"https://openalex.org/C2775997480","wikidata":"https://www.wikidata.org/wiki/Q586277","display_name":"Degree (music)","level":2,"score":0.35040000081062317},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3149999976158142},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.26420000195503235}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.05464","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05464","pdf_url":"https://arxiv.org/pdf/2608.05464","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.2608.05464","doi":"https://doi.org/10.48550/arxiv.2608.05464","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.05464","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:2608.05464","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.05464","pdf_url":"https://arxiv.org/pdf/2608.05464","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":[{"id":"https://openalex.org/G1021461588","display_name":null,"funder_award_id":"FA 9550-22-1-0532","funder_id":"https://openalex.org/F4320338279","funder_display_name":"Air Force Office of Scientific Research"},{"id":"https://openalex.org/G1523888516","display_name":null,"funder_award_id":"FA9550-","funder_id":"https://openalex.org/F4320338279","funder_display_name":"Air Force Office of Scientific Research"}],"funders":[{"id":"https://openalex.org/F4320338279","display_name":"Air Force Office of Scientific Research","ror":"https://ror.org/011e9bt93"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7201870942.pdf","grobid_xml":"https://content.openalex.org/works/W7201870942.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W1980789245","https://openalex.org/W2083252150","https://openalex.org/W2255138985","https://openalex.org/W2782973718","https://openalex.org/W2964121744","https://openalex.org/W3016391357","https://openalex.org/W3037787879","https://openalex.org/W4401726555","https://openalex.org/W4402810885"],"related_works":["https://openalex.org/W3106352472","https://openalex.org/W7163201870","https://openalex.org/W6941089071","https://openalex.org/W4281706755","https://openalex.org/W7030482579","https://openalex.org/W2153827974","https://openalex.org/W2991340425","https://openalex.org/W7212162542","https://openalex.org/W2071676111","https://openalex.org/W3207964041"],"abstract_inverted_index":{"Pruning":[0],"network":[1],"connections":[2,93,101,111],"is":[3,123,144],"key":[4],"to":[5,39,102,112,125],"brain":[6],"function":[7],"but,":[8],"despite":[9],"its":[10],"importance,":[11],"there":[12],"exist":[13],"few":[14],"biologically-plausible":[15],"pruning":[16,30,132,172],"rules":[17],"with":[18,193],"demonstrated":[19],"good":[20,126],"performance.":[21],"We":[22,60,116,134],"evaluate":[23],"noise-prune,":[24],"a":[25,56,73,82,97,156,170],"recently":[26],"introduced":[27],"unsupervised":[28],"local":[29,176],"rule":[31],"for":[32],"recurrent":[33,68,186],"networks":[34,54,187],"that":[35,62,75,85,94,118,137,161,169],"uses":[36,77,86],"noisy":[37],"fluctuations":[38],"determine":[40],"the":[41,138,149,163],"importance":[42,107],"of":[43,142,159,196],"connections.":[44,198],"Noise-prune":[45],"has":[46],"previously":[47],"only":[48,76],"been":[49],"empirically":[50],"tested":[51],"on":[52,105,175],"random":[53],"without":[55],"specific":[57],"computational":[58],"function.":[59],"show":[61,117],"noise-prune":[63,99,160],"preserves":[64],"task-performance":[65],"in":[66,184],"task-trained":[67],"neural":[69],"networks,":[70],"greatly":[71],"outperforming":[72],"strategy":[74,84,173],"connection":[78,191],"magnitude":[79],"and":[80,108,121,128,153],"exceeding":[81],"non-local":[83],"second-order":[87],"information.":[88],"Rather":[89],"than":[90,146],"deterministically":[91],"removing":[92],"fall":[95],"below":[96],"threshold,":[98],"samples":[100],"preserve":[103,113,181],"based":[104,174],"their":[106],"strengthens":[109],"retained":[110],"average":[114],"strength.":[115],"this":[119],"sampling":[120],"rescaling":[122,143],"essential":[124],"performance,":[127],"can":[129,179],"augment":[130],"other":[131],"strategies.":[133],"also":[135],"find":[136],"optimal":[139],"empirical":[140],"degree":[141],"lower":[145],"predicted":[147],"by":[148,188],"original":[150],"theoretical":[151],"argument,":[152],"thus":[154],"develop":[155],"new":[157],"version":[158],"outperforms":[162],"previous":[164],"version.":[165],"Our":[166],"results":[167],"demonstrate":[168],"biologically-motivated":[171],"activity":[177],"statistics":[178],"effectively":[180],"task":[182],"performance":[183],"functional":[185],"combining":[189],"activity-dependent":[190],"selection":[192],"compensatory":[194],"strengthening":[195],"surviving":[197]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-08-08T00:00:00"}