{"id":"https://openalex.org/W4405215425","doi":"https://doi.org/10.32388/xlqw1l","title":"Proactive Gradient Conflict Mitigation in Multi-Task Learning:A Sparse Training Perspective","display_name":"Proactive Gradient Conflict Mitigation in Multi-Task Learning:A Sparse Training Perspective","publication_year":2024,"publication_date":"2024-12-10","ids":{"openalex":"https://openalex.org/W4405215425","doi":"https://doi.org/10.32388/xlqw1l"},"language":"en","primary_location":{"id":"doi:10.32388/xlqw1l","doi":"https://doi.org/10.32388/xlqw1l","is_oa":true,"landing_page_url":"https://doi.org/10.32388/xlqw1l","pdf_url":"https://www.qeios.com/read/XLQW1L/pdf","source":{"id":"https://openalex.org/S4210177074","display_name":"Qeios","issn_l":"2632-3834","issn":["2632-3834"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["anvur-scientific","norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Qeios","raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://www.qeios.com/read/XLQW1L/pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100749614","display_name":"Zhang Zhi","orcid":"https://orcid.org/0000-0002-1476-6790","observed_orcids":["https://orcid.org/0000-0002-1476-6790","https://orcid.org/0000-0002-9817-9488"]},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Zhi Zhang","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100766009","display_name":"Jiayi Shen","orcid":"https://orcid.org/0000-0002-0012-1457","observed_orcids":["https://orcid.org/0000-0002-0012-1457","https://orcid.org/0000-0002-6348-2559"]},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Jiayi Shen","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013030532","display_name":"Shanghang Zhang","orcid":"https://orcid.org/0000-0003-4047-3526","observed_orcids":["https://orcid.org/0000-0003-4047-3526"]},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shanghang Zhang","raw_affiliation_strings":["Peking University"],"raw_orcid":"https://orcid.org/0000-0003-4047-3526","affiliations":[{"raw_affiliation_string":"Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016184654","display_name":"Ekaterina Shutova","orcid":"https://orcid.org/0009-0003-6664-4474","observed_orcids":["https://orcid.org/0009-0003-6664-4474"]},"institutions":[{"id":"https://openalex.org/I887064364","display_name":"University of Amsterdam","ror":"https://ror.org/04dkp9463","country_code":"NL","type":"education","lineage":["https://openalex.org/I887064364"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Ekaterina Shutova","raw_affiliation_strings":["University of Amsterdam"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Amsterdam","institution_ids":["https://openalex.org/I887064364"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":4,"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":0.98580002784729,"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":0.98580002784729,"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":0.005400000140070915,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.002199999988079071,"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/gradient-conflict","display_name":"gradient conflict","score":0.8669999837875366},{"id":"https://openalex.org/keywords/multi-task-learning","display_name":"multi-task learning","score":0.996999979019165},{"id":"https://openalex.org/keywords/sparse-training","display_name":"sparse training","score":0.996999979019165},{"id":"https://openalex.org/keywords/task-balancing","display_name":"task balancing","score":0.6190000176429749},{"id":"https://openalex.org/keywords/parameter-update","display_name":"parameter updating","score":0.3970000147819519}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7672773003578186},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6909835934638977},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.6773066520690918},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.5842224955558777},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4657045006752014},{"id":"https://openalex.org/C91306197","wikidata":"https://www.wikidata.org/wiki/Q45767","display_name":"Competition (biology)","level":2,"score":0.4169505834579468},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37971144914627075},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08091658353805542},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.0},{"id":"https://openalex.org/C153294291","wikidata":"https://www.wikidata.org/wiki/Q25261","display_name":"Meteorology","level":1,"score":0.0},{"id":"https://openalex.org/C18903297","wikidata":"https://www.wikidata.org/wiki/Q7150","display_name":"Ecology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.32388/xlqw1l","doi":"https://doi.org/10.32388/xlqw1l","is_oa":true,"landing_page_url":"https://doi.org/10.32388/xlqw1l","pdf_url":"https://www.qeios.com/read/XLQW1L/pdf","source":{"id":"https://openalex.org/S4210177074","display_name":"Qeios","issn_l":"2632-3834","issn":["2632-3834"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["anvur-scientific","norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Qeios","raw_type":"posted-content"},{"id":"pmh:oai:arXiv.org:2411.18615","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2411.18615","pdf_url":"https://arxiv.org/pdf/2411.18615","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.2411.18615","doi":"https://doi.org/10.48550/arxiv.2411.18615","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2411.18615","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":"doi:10.32388/xlqw1l","doi":"https://doi.org/10.32388/xlqw1l","is_oa":true,"landing_page_url":"https://doi.org/10.32388/xlqw1l","pdf_url":"https://www.qeios.com/read/XLQW1L/pdf","source":{"id":"https://openalex.org/S4210177074","display_name":"Qeios","issn_l":"2632-3834","issn":["2632-3834"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":["anvur-scientific","norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Qeios","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4405215425.pdf","grobid_xml":"https://content.openalex.org/works/W4405215425.grobid-xml"},"referenced_works_count":38,"referenced_works":["https://openalex.org/W1834627138","https://openalex.org/W1910657905","https://openalex.org/W2108598243","https://openalex.org/W2147062835","https://openalex.org/W2339391301","https://openalex.org/W2340897893","https://openalex.org/W2798441115","https://openalex.org/W2805003733","https://openalex.org/W2890538051","https://openalex.org/W2895401575","https://openalex.org/W2936484090","https://openalex.org/W2959581809","https://openalex.org/W2963038646","https://openalex.org/W2963430933","https://openalex.org/W2963677766","https://openalex.org/W2997359900","https://openalex.org/W2998103904","https://openalex.org/W3008526508","https://openalex.org/W3024171804","https://openalex.org/W3033503043","https://openalex.org/W3085046840","https://openalex.org/W3094502228","https://openalex.org/W3105787868","https://openalex.org/W3129958428","https://openalex.org/W3138516171","https://openalex.org/W3141797743","https://openalex.org/W3154435685","https://openalex.org/W3199132546","https://openalex.org/W3199761064","https://openalex.org/W3203399909","https://openalex.org/W3209347846","https://openalex.org/W4225981399","https://openalex.org/W4321649543","https://openalex.org/W4382202657","https://openalex.org/W4390190100","https://openalex.org/W4390874575","https://openalex.org/W4402703104","https://openalex.org/W7133184305"],"related_works":["https://openalex.org/W4312274013","https://openalex.org/W4402381588","https://openalex.org/W3209347846","https://openalex.org/W4321649543","https://openalex.org/W4386938047","https://openalex.org/W4407953228","https://openalex.org/W4295141442","https://openalex.org/W4400909580","https://openalex.org/W4413156745","https://openalex.org/W4416644612"],"abstract_inverted_index":{"Advancing":[0],"towards":[1],"generalist":[2],"agents":[3],"necessitates":[4],"the":[5,17,35,61,89,100,125,134],"concurrent":[6],"processing":[7],"of":[8,20,37,63,91,102,124],"multiple":[9,25],"tasks":[10,47],"using":[11],"a":[12,110,122],"unified":[13],"model,":[14],"thereby":[15],"underscoring":[16],"growing":[18],"significance":[19],"simultaneous":[21],"model":[22],"training":[23,118,131],"on":[24],"downstream":[26],"tasks.":[27],"A":[28],"common":[29],"issue":[30,77],"in":[31,55,57,65],"multi-task":[32],"learning":[33],"is":[34],"occurrence":[36,101],"gradient":[38,92,103,159],"conflict,":[39],"which":[40],"leads":[41,148],"to":[42,74,112,149],"potential":[43],"competition":[44,52],"among":[45],"different":[46,106],"during":[48,130],"joint":[49],"training.":[50],"This":[51],"often":[53],"results":[54],"improvements":[56],"one":[58],"task":[59,80,84],"at":[60],"expense":[62],"deterioration":[64],"another.":[66],"Although":[67],"several":[68],"optimization":[69],"methods":[70,107],"have":[71],"been":[72],"developed":[73],"address":[75],"this":[76,95],"by":[78],"manipulating":[79],"gradients":[81,146],"for":[82],"better":[83],"balancing,":[85],"they":[86],"cannot":[87],"decrease":[88],"incidence":[90],"conflict.":[93],"In":[94],"paper,":[96],"we":[97],"systematically":[98],"investigate":[99],"conflict":[104],"across":[105],"and":[108,147],"propose":[109],"strategy":[111],"reduce":[113],"such":[114],"conflicts":[115],"through":[116],"sparse":[117],"(ST),":[119],"wherein":[120],"only":[121],"portion":[123],"model\u2019s":[126],"parameters":[127],"are":[128],"updated":[129],"while":[132],"keeping":[133],"rest":[135],"unchanged.":[136],"Our":[137],"extensive":[138],"experiments":[139],"demonstrate":[140],"that":[141],"ST":[142,153],"effectively":[143],"mitigates":[144],"conflicting":[145],"superior":[150],"performance.":[151],"Furthermore,":[152],"can":[154],"be":[155],"easily":[156],"integrated":[157],"with":[158],"manipulation":[160],"techniques,":[161],"thus":[162],"enhancing":[163],"their":[164],"effectiveness.":[165]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2024-12-11T00:00:00"}