{"id":"https://openalex.org/W3132646476","doi":"https://doi.org/10.48550/arxiv.2010.04495","title":"Linear Mode Connectivity in Multitask and Continual Learning","display_name":"Linear Mode Connectivity in Multitask and Continual Learning","publication_year":2020,"publication_date":"2020-10-09","ids":{"openalex":"https://openalex.org/W3132646476","doi":"https://doi.org/10.48550/arxiv.2010.04495","mag":"3132646476"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2010.04495","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2010.04495","pdf_url":"https://arxiv.org/pdf/2010.04495","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/2010.04495","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010129552","display_name":"Seyed Iman Mirzadeh","orcid":"https://orcid.org/0000-0003-1586-0700","observed_orcids":["https://orcid.org/0000-0003-1586-0700"]},"institutions":[{"id":"https://openalex.org/I72951846","display_name":"Washington State University","ror":"https://ror.org/05dk0ce17","country_code":"US","type":"education","lineage":["https://openalex.org/I72951846"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mirzadeh, Seyed Iman","raw_affiliation_strings":["washington state university"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"washington state university","institution_ids":["https://openalex.org/I72951846"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050499655","display_name":"Mehrdad Farajtabar","orcid":"https://orcid.org/0000-0002-5510-518X","observed_orcids":["https://orcid.org/0000-0002-5510-518X"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Farajtabar, Mehrdad","raw_affiliation_strings":["DeepMind , USA","Google,,,,,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"DeepMind , USA","institution_ids":[]},{"raw_affiliation_string":"Google,,,,,","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081258844","display_name":"Dilan G\u00f6r\u00fcr","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gorur, Dilan","raw_affiliation_strings":["DeepMind , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"DeepMind , USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043910056","display_name":"Razvan Pascanu","orcid":"https://orcid.org/0000-0002-5470-1238","observed_orcids":["https://orcid.org/0000-0002-5470-1238"]},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]},{"id":"https://openalex.org/I4210090411","display_name":"Google DeepMind (United Kingdom)","ror":"https://ror.org/00971b260","country_code":"GB","type":"company","lineage":["https://openalex.org/I4210090411","https://openalex.org/I4210128969"]}],"countries":["GB","US"],"is_corresponding":false,"raw_author_name":"Pascanu, Razvan","raw_affiliation_strings":["DeepMind , UK","Google DeepMind;"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"DeepMind , UK","institution_ids":["https://openalex.org/I4210090411"]},{"raw_affiliation_string":"Google DeepMind;","institution_ids":["https://openalex.org/I1291425158","https://openalex.org/I4210090411"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007139473","display_name":"Hassan G Ghasemzadeh","orcid":"https://orcid.org/0000-0002-1844-1416","observed_orcids":["https://orcid.org/0000-0002-1844-1416"]},"institutions":[{"id":"https://openalex.org/I72951846","display_name":"Washington State University","ror":"https://ror.org/05dk0ce17","country_code":"US","type":"education","lineage":["https://openalex.org/I72951846"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ghasemzadeh, Hassan","raw_affiliation_strings":["Washington State University , USA","Washington State University , Pullman"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Washington State University , USA","institution_ids":["https://openalex.org/I72951846"]},{"raw_affiliation_string":"Washington State University , Pullman","institution_ids":["https://openalex.org/I72951846"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":95,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9970999956130981,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9970999956130981,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0003000000142492354,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":9.999999747378752e-05,"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/continual-learning","display_name":"continual learning","score":0.800000011920929},{"id":"https://openalex.org/keywords/multi-task-learning","display_name":"multi-task learning","score":0.9639999866485596},{"id":"https://openalex.org/keywords/catastrophic-forgetting","display_name":"catastrophic forgetting","score":0.9079999923706055},{"id":"https://openalex.org/keywords/loss-landscape","display_name":"loss landscape","score":0.3580000102519989}],"concepts":[{"id":"https://openalex.org/C186633575","wikidata":"https://www.wikidata.org/wiki/Q845060","display_name":"Maxima and minima","level":2,"score":0.8168221116065979},{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.8027437925338745},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7656744718551636},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.621324360370636},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6174270510673523},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.6091822385787964},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.6070773005485535},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5960419178009033},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5690356492996216},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5004630088806152},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4851280152797699},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.46987301111221313},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.41804784536361694},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.12397325038909912},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2010.04495","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2010.04495","pdf_url":"https://arxiv.org/pdf/2010.04495","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.2010.04495","doi":"https://doi.org/10.48550/arxiv.2010.04495","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2010.04495","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"},{"id":"mag:3132646476","doi":null,"is_oa":false,"landing_page_url":"https://openreview.net/pdf?id=Fmg_fQYUejf","pdf_url":null,"source":null,"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":null}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2010.04495","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2010.04495","pdf_url":"https://arxiv.org/pdf/2010.04495","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/G1930397893","display_name":"CAREER: Autonomous Wearable Computing for Personalized Healthcare","funder_award_id":"1750679","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W3132646476.pdf","grobid_xml":"https://content.openalex.org/works/W3132646476.grobid-xml"},"referenced_works_count":74,"referenced_works":["https://openalex.org/W1682403713","https://openalex.org/W2113839990","https://openalex.org/W2116522068","https://openalex.org/W2426267443","https://openalex.org/W2554863749","https://openalex.org/W2560647685","https://openalex.org/W2605043629","https://openalex.org/W2618767506","https://openalex.org/W2737492962","https://openalex.org/W2738110777","https://openalex.org/W2743151379","https://openalex.org/W2744654301","https://openalex.org/W2764043458","https://openalex.org/W2768412495","https://openalex.org/W2785567898","https://openalex.org/W2786446225","https://openalex.org/W2787295326","https://openalex.org/W2788388592","https://openalex.org/W2793333878","https://openalex.org/W2798836702","https://openalex.org/W2803552920","https://openalex.org/W2804746922","https://openalex.org/W2809090039","https://openalex.org/W2895699061","https://openalex.org/W2898290017","https://openalex.org/W2899063268","https://openalex.org/W2899320936","https://openalex.org/W2902456977","https://openalex.org/W2902625698","https://openalex.org/W2903183390","https://openalex.org/W2903996579","https://openalex.org/W2912515466","https://openalex.org/W2918932727","https://openalex.org/W2922031930","https://openalex.org/W2924984511","https://openalex.org/W2932484476","https://openalex.org/W2942810103","https://openalex.org/W2945158659","https://openalex.org/W2947461406","https://openalex.org/W2948376147","https://openalex.org/W2950184636","https://openalex.org/W2950233433","https://openalex.org/W2952571470","https://openalex.org/W2962724315","https://openalex.org/W2963384892","https://openalex.org/W2963577698","https://openalex.org/W2963674932","https://openalex.org/W2963850662","https://openalex.org/W2964790801","https://openalex.org/W2965849096","https://openalex.org/W2970239942","https://openalex.org/W2970502342","https://openalex.org/W2970586779","https://openalex.org/W2970971581","https://openalex.org/W2971184120","https://openalex.org/W2973737673","https://openalex.org/W2979646054","https://openalex.org/W2980753060","https://openalex.org/W2992525328","https://openalex.org/W2996653965","https://openalex.org/W3000677656","https://openalex.org/W3005730716","https://openalex.org/W3020401063","https://openalex.org/W3030364939","https://openalex.org/W3035081900","https://openalex.org/W3036362369","https://openalex.org/W3037853434","https://openalex.org/W3040607201","https://openalex.org/W3043058418","https://openalex.org/W3090618925","https://openalex.org/W3094105627","https://openalex.org/W3098033339","https://openalex.org/W3103936143","https://openalex.org/W4319988532"],"related_works":["https://openalex.org/W4319659583","https://openalex.org/W4415799122","https://openalex.org/W7116040620","https://openalex.org/W7128028851","https://openalex.org/W4312801469","https://openalex.org/W4389115821","https://openalex.org/W2964790801","https://openalex.org/W4307478737","https://openalex.org/W4320351890","https://openalex.org/W7006886796"],"abstract_inverted_index":{"Continual":[0],"(sequential)":[1],"training":[2,6,33,85],"and":[3,141,182],"multitask":[4,140,211],"(simultaneous)":[5],"are":[7,127,144],"often":[8],"attempting":[9],"to":[10,16,42,54,196,207],"solve":[11],"the":[12,32,59,71,76,79,83,105,111,124,172,187,203,210,222],"same":[13,125,173],"overall":[14],"objective:":[15],"find":[17,149],"a":[18,46,63,99,166],"solution":[19,60],"that":[20,82,102,120,150,201,215],"performs":[21],"well":[22,69,91],"on":[23,70,170,227],"all":[24],"considered":[25],"tasks.":[26],"The":[27],"main":[28],"difference":[29,106],"is":[30,89],"in":[31,107],"regimes,":[34],"where":[35],"continual":[36,142,188,224],"learning":[37,189,225],"can":[38,154,162],"only":[39],"have":[40],"access":[41],"one":[43],"task":[44,65,126],"at":[45,88],"time,":[47],"which":[48],"for":[49,62,175,186],"neural":[50],"networks":[51],"typically":[52,128],"leads":[53],"catastrophic":[55],"forgetting.":[56],"That":[57],"is,":[58],"found":[61],"subsequent":[64],"does":[66],"not":[67,90],"perform":[68],"previous":[72],"ones":[73],"anymore.":[74],"However,":[75],"relationship":[77],"between":[78],"different":[80,113,121],"minima":[81,122,206],"two":[84,112],"regimes":[86],"arrive":[87],"understood.":[92],"What":[93],"sets":[94],"them":[95],"apart?":[96],"Is":[97],"there":[98],"local":[100],"structure":[101],"could":[103],"explain":[104],"performance":[108],"achieved":[109,157],"by":[110,116,130,165],"schemes?":[114],"Motivated":[115],"recent":[117],"work":[118],"showing":[119],"of":[123,134,221],"connected":[129],"very":[131],"simple":[132],"curves":[133],"low":[135],"error,":[136],"we":[137,192],"investigate":[138],"whether":[139],"solutions":[143],"similarly":[145],"connected.":[146],"We":[147,177,213],"empirically":[148],"indeed":[151],"such":[152],"connectivity":[153],"be":[155,163],"reliably":[156],"and,":[158],"more":[159],"interestingly,":[160],"it":[161],"done":[164],"linear":[167],"path,":[168],"conditioned":[169],"having":[171],"initialization":[174],"both.":[176],"thoroughly":[178],"analyze":[179],"this":[180,194],"observation":[181],"discuss":[183],"its":[184],"significance":[185],"process.":[190],"Furthermore,":[191],"exploit":[193],"finding":[195],"propose":[197],"an":[198],"effective":[199],"algorithm":[200],"constrains":[202],"sequentially":[204],"learned":[205],"behave":[208],"as":[209],"solution.":[212],"show":[214],"our":[216],"method":[217],"outperforms":[218],"several":[219],"state":[220],"art":[223],"algorithms":[226],"various":[228],"vision":[229],"benchmarks.":[230]},"counts_by_year":[{"year":2026,"cited_by_count":16},{"year":2025,"cited_by_count":14},{"year":2024,"cited_by_count":26},{"year":2023,"cited_by_count":17},{"year":2022,"cited_by_count":10},{"year":2021,"cited_by_count":12}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2021-03-01T00:00:00"}