{"id":"https://openalex.org/W7225457167","doi":"https://doi.org/10.5281/zenodo.24032344","title":"Phase learning-rate multiplier in a complex recurrent cell","display_name":"Phase learning-rate multiplier in a complex recurrent cell","publication_year":2026,"publication_date":"2026-11-07","ids":{"openalex":"https://openalex.org/W7225457167","doi":"https://doi.org/10.5281/zenodo.24032344"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24032344","doi":"https://doi.org/10.5281/zenodo.24032344","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24032344","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24032344","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161172933","display_name":"Ziedonis Valbergs","orcid":"https://orcid.org/0000-0003-3352-273X","observed_orcids":["https://orcid.org/0000-0003-3352-273X"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Valbergs, Ziedonis","raw_affiliation_strings":["Aura Mind"],"raw_orcid":"https://orcid.org/0009-0000-1442-0629","affiliations":[{"raw_affiliation_string":"Aura Mind","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5161172933"],"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/T10028","display_name":"Topic Modeling","score":0.7631999850273132,"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/T10028","display_name":"Topic Modeling","score":0.7631999850273132,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.0778999999165535,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.029500000178813934,"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/learning-rate-scaling","display_name":"learning rate scaling","score":0.7110000252723694},{"id":"https://openalex.org/keywords/recurrent-neural-networks","display_name":"recurrent neural networks","score":0.6430000066757202},{"id":"https://openalex.org/keywords/language-model-training","display_name":"language model training","score":0.4490000009536743},{"id":"https://openalex.org/keywords/cross-lingual-evaluation","display_name":"cross-lingual evaluation","score":0.5770000219345093},{"id":"https://openalex.org/keywords/biomedical-text","display_name":"biomedical text","score":0.9520000219345093},{"id":"https://openalex.org/keywords/training-stability","display_name":"training stability","score":0.34700000286102295}],"concepts":[{"id":"https://openalex.org/C109167261","wikidata":"https://www.wikidata.org/wiki/Q5421955","display_name":"Fell","level":2,"score":0.7972000241279602},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5618000030517578},{"id":"https://openalex.org/C94375191","wikidata":"https://www.wikidata.org/wiki/Q11205","display_name":"Arithmetic","level":1,"score":0.5128999948501587},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.5059999823570251},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.5015000104904175},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.48249998688697815},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.3869999945163727},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.38600000739097595},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.3799000084400177},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3732999861240387},{"id":"https://openalex.org/C44280652","wikidata":"https://www.wikidata.org/wiki/Q104837","display_name":"Phase (matter)","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.32690000534057617},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31200000643730164},{"id":"https://openalex.org/C2779079380","wikidata":"https://www.wikidata.org/wiki/Q333495","display_name":"Falling (accident)","level":2,"score":0.29739999771118164},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.2953999936580658},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.29269999265670776},{"id":"https://openalex.org/C2780813799","wikidata":"https://www.wikidata.org/wiki/Q3274237","display_name":"Zero (linguistics)","level":2,"score":0.2822999954223633},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2578999996185303},{"id":"https://openalex.org/C2987804222","wikidata":"https://www.wikidata.org/wiki/Q5376582","display_name":"End point","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.5281/zenodo.24032344","doi":"https://doi.org/10.5281/zenodo.24032344","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24032344","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},{"id":"doi:10.5281/zenodo.24032345","doi":"https://doi.org/10.5281/zenodo.24032345","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24032345","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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"},{"id":"doi:10.5281/zenodo.24032876","doi":"https://doi.org/10.5281/zenodo.24032876","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24032876","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"doi:10.5281/zenodo.24032344","doi":"https://doi.org/10.5281/zenodo.24032344","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24032344","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7155565028","https://openalex.org/W3123178201","https://openalex.org/W7214405976","https://openalex.org/W7116123065","https://openalex.org/W3154976557","https://openalex.org/W2466579669","https://openalex.org/W7127923911","https://openalex.org/W7197016094","https://openalex.org/W2898501133","https://openalex.org/W4403964581"],"abstract_inverted_index":{"Multiplying":[0],"the":[1,5,47,74,89,129,132,173],"optimizer":[2],"step":[3],"of":[4,87,126],"phase":[6],"parameters":[7],"by":[8,153],"10":[9],"to":[10,33,41,58,70,111,115,137,146],"80":[11],"lowered":[12],"loss":[13,51],"onunseen":[14],"text":[15,99],"in":[16,93,117],"a":[17,36,124,187],"complex":[18],"recurrent":[19],"cell,":[20],"at":[21,78,141,186],"no":[22],"measurable":[23],"compute":[24],"cost.":[25],"We":[26],"trained":[27,180],"a27M-parameter":[28],"model":[29,38,182],"(2000":[30],"steps,":[31],"\u00d71":[32,116],"\u00d7160)":[34],"and":[35,64,81,97,143,155,167],"101M-parameter":[37],"(4000":[39],"steps,\u00d71":[40],"\u00d7320),":[42],"with":[43,190],"three":[44,95],"seeds":[45,151],"each.In":[46],"100M":[48],"model,":[49,131],"mean":[50],"on":[52,65,164],"unseen":[53,66],"English":[54,80],"fell":[55,113,134],"from":[56,68,135],"3.807":[57],"3.773":[59],"nats":[60],"per":[61,199],"token":[62],"at\u00d780,":[63],"Latvian":[67,82],"3.330":[69],"3.295.":[71],"Beyond":[72],"\u00d780":[73],"gain":[75],"stopped":[76],"growingreliably:":[77],"\u00d7320":[79,112],"stayed":[83,144],"within":[84,196],"seed":[85],"spread":[86,152],"\u00d780,":[88],"trainingmonitor":[90],"was":[91,194],"worse":[92],"all":[94],"seeds,":[96],"biomedical":[98],"kept":[100],"falling":[101],"while":[102],"its":[103,150,159],"seedsspread":[104],"3.6":[105],"times":[106],"wider.":[107],"No":[108],"multiplier":[109],"up":[110],"back":[114],"any":[118],"seed,":[119],"so":[120],"we":[121],"did":[122],"notreach":[123],"point":[125],"failure.":[127],"In":[128],"small":[130,181],"monitor":[133],"1.4931":[136],"1.4840":[138],"bits":[139,198],"perbyte":[140],"\u00d710":[142,193],"flat":[145],"\u00d780.":[147],"At":[148],"\u00d7160":[149],"0.0168":[154],"one":[156,165],"endedworse":[157],"than":[158],"baseline.These":[160],"are":[161],"early-training":[162],"results":[163],"architecture,":[166],"they":[168],"do":[169],"not":[170],"show":[171],"that":[172],"gainsurvives":[174],"long":[175],"training.":[176],"Continuing":[177],"an":[178],"already":[179],"for":[183],"1000":[184],"steps":[185],"lowlearning":[188],"rate":[189],"\u00d73":[191],"or":[192],"neutral,":[195],"0.0001":[197],"byte":[200]},"counts_by_year":[],"updated_date":"2026-11-11T12:16:26.678958","created_date":"2026-11-08T00:00:00"}