{"id":"https://openalex.org/W7168300087","doi":"https://doi.org/10.48550/arxiv.2607.11555","title":"Advancing Optimal Subset Oracle via Learning Relaxation of Neural Set Functions","display_name":"Advancing Optimal Subset Oracle via Learning Relaxation of Neural Set Functions","publication_year":2026,"publication_date":"2026-07-13","ids":{"openalex":"https://openalex.org/W7168300087","doi":"https://doi.org/10.48550/arxiv.2607.11555"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2607.11555","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.11555","pdf_url":"https://arxiv.org/pdf/2607.11555","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/2607.11555","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140573218","display_name":"Yongquan Shi","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I80947539","display_name":"Fuzhou University","ror":"https://ror.org/011xvna82","country_code":"CN","type":"education","lineage":["https://openalex.org/I80947539"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shi, Yongquan","raw_affiliation_strings":["Fuzhou University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fuzhou University ,","institution_ids":["https://openalex.org/I80947539"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021091982","display_name":"Zijing Ou","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47508984","display_name":"Imperial College London","ror":"https://ror.org/041kmwe10","country_code":"GB","type":"education","lineage":["https://openalex.org/I47508984"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ou, Zijing","raw_affiliation_strings":["Imperial College London ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Imperial College London ,","institution_ids":["https://openalex.org/I47508984"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125750837","display_name":"Shiping Wang","orcid":"https://orcid.org/0000-0001-5195-9682","observed_orcids":["https://orcid.org/0000-0001-5195-9682"]},"institutions":[{"id":"https://openalex.org/I80947539","display_name":"Fuzhou University","ror":"https://ror.org/011xvna82","country_code":"CN","type":"education","lineage":["https://openalex.org/I80947539"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Shiping","raw_affiliation_strings":["Fuzhou University ,"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Fuzhou University ,","institution_ids":["https://openalex.org/I80947539"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045777220","display_name":"Yatao Bian","orcid":"https://orcid.org/0000-0002-2368-4084","observed_orcids":["https://orcid.org/0000-0002-2368-4084"]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Bian, Yatao","raw_affiliation_strings":["National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]}],"institutions":[],"countries_distinct_count":3,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.12809999287128448,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.12809999287128448,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.06260000169277191,"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/T10203","display_name":"Recommender Systems and Techniques","score":0.043299999088048935,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/weakly-supervised-learning","display_name":"weakly supervised learning","score":0.7039999961853027},{"id":"https://openalex.org/keywords/mean-field-variational-inference","display_name":"mean-field variational inference","score":0.875},{"id":"https://openalex.org/keywords/evidence-lower-bound","display_name":"evidence lower bound","score":0.9150000214576721},{"id":"https://openalex.org/keywords/continuous-relaxation","display_name":"continuous relaxation","score":0.8330000042915344},{"id":"https://openalex.org/keywords/surrogate-objective","display_name":"surrogate objective","score":0.8790000081062317},{"id":"https://openalex.org/keywords/submodular-maximization","display_name":"submodular maximization","score":0.8169999718666077},{"id":"https://openalex.org/keywords/variational-free-energy","display_name":"variational free energy","score":0.9909999966621399},{"id":"https://openalex.org/keywords/compound-selection","display_name":"compound selection","score":0.925000011920929},{"id":"https://openalex.org/keywords/drug-discovery","display_name":"drug discovery","score":0.7979999780654907},{"id":"https://openalex.org/keywords/item-recommendation","display_name":"product recommendation","score":0.7829999923706055}],"concepts":[{"id":"https://openalex.org/C178621042","wikidata":"https://www.wikidata.org/wiki/Q7631710","display_name":"Submodular set function","level":2,"score":0.8027999997138977},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.6870999932289124},{"id":"https://openalex.org/C2776029896","wikidata":"https://www.wikidata.org/wiki/Q3935810","display_name":"Relaxation (psychology)","level":2,"score":0.5684000253677368},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.5652999877929688},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5321000218391418},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5277000069618225},{"id":"https://openalex.org/C2776330181","wikidata":"https://www.wikidata.org/wiki/Q18358244","display_name":"Maximization","level":2,"score":0.48249998688697815},{"id":"https://openalex.org/C15312841","wikidata":"https://www.wikidata.org/wiki/Q1657198","display_name":"Set function","level":3,"score":0.4657000005245209},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.44530001282691956},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.42910000681877136},{"id":"https://openalex.org/C77553402","wikidata":"https://www.wikidata.org/wiki/Q13222579","display_name":"Upper and lower bounds","level":2,"score":0.42080000042915344},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.413100004196167},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4101000130176544},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3864000141620636},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.37940001487731934},{"id":"https://openalex.org/C19499675","wikidata":"https://www.wikidata.org/wiki/Q232207","display_name":"Monte Carlo method","level":2,"score":0.36399999260902405},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.35420000553131104},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.33660000562667847},{"id":"https://openalex.org/C13355873","wikidata":"https://www.wikidata.org/wiki/Q2920850","display_name":"Connection (principal bundle)","level":2,"score":0.33649998903274536},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.335999995470047},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3167000114917755},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31139999628067017},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.28600001335144043},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.28209999203681946},{"id":"https://openalex.org/C131675550","wikidata":"https://www.wikidata.org/wiki/Q7646884","display_name":"Surrogate model","level":2,"score":0.2791000008583069},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C91873725","wikidata":"https://www.wikidata.org/wiki/Q3445816","display_name":"Function approximation","level":3,"score":0.2653999924659729},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.25920000672340393}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2607.11555","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.11555","pdf_url":"https://arxiv.org/pdf/2607.11555","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.2607.11555","doi":"https://doi.org/10.48550/arxiv.2607.11555","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.11555","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:2607.11555","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2607.11555","pdf_url":"https://arxiv.org/pdf/2607.11555","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/W7168300087.pdf","grobid_xml":"https://content.openalex.org/works/W7168300087.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7133205896","https://openalex.org/W7133196101","https://openalex.org/W2918466559","https://openalex.org/W4391622580","https://openalex.org/W2947798871","https://openalex.org/W3006403619","https://openalex.org/W7212326948","https://openalex.org/W2944971886","https://openalex.org/W6948160945","https://openalex.org/W3138674686"],"abstract_inverted_index":{"Learning":[0],"neural":[1],"set":[2,33,102],"functions":[3,34],"is":[4],"pivotal":[5],"to":[6,30,57,154],"a":[7,97,106,160],"wide":[8],"range":[9],"of":[10,60,100,162],"important":[11],"applications,":[12],"including":[13],"compound":[14],"selection":[15],"in":[16],"AI-driven":[17],"drug":[18],"discovery":[19],"and":[20,104,123,134,150],"product":[21],"recommendation.":[22],"Recent":[23],"work":[24],"has":[25],"introduced":[26],"optimal":[27],"subset":[28],"oracles":[29],"implicitly":[31],"learn":[32,105],"under":[35,147],"practical":[36],"weakly":[37],"supervised":[38],"settings,":[39],"where":[40],"model":[41],"parameters":[42],"are":[43],"optimized":[44],"through":[45],"mean-field":[46],"variational":[47,68,116,155],"inference.":[48,136],"However,":[49],"these":[50],"frameworks":[51],"rely":[52],"on":[53,159],"Monte":[54],"Carlo":[55],"sampling":[56,71],"estimate":[58],"gradients":[59,125],"the":[61,67,79,84,92,101,127,144],"evidence":[62,93],"lower":[63,94],"bound":[64,95],"when":[65],"updating":[66],"distribution.":[69],"Repeated":[70],"across":[72],"iterations":[73],"incurs":[74],"substantial":[75],"computational":[76,132],"overhead,":[77],"while":[78],"resulting":[80],"stochasticity":[81],"can":[82],"destabilize":[83],"optimization":[85],"trajectory.":[86],"In":[87],"this":[88],"work,":[89],"we":[90,138],"reinterpret":[91],"as":[96],"continuous":[98,128],"relaxation":[99],"function":[103],"surrogate":[107,120],"objective":[108],"that":[109],"replaces":[110],"sampling-based":[111],"ELBO":[112],"gradient":[113],"estimation":[114],"during":[115],"optimization.":[117],"The":[118],"learned":[119],"provides":[121],"stable":[122],"efficient":[124],"throughout":[126],"domain,":[129],"thereby":[130],"reducing":[131],"overhead":[133],"accelerating":[135],"Furthermore,":[137],"establish":[139],"an":[140],"approximation":[141],"guarantee":[142],"for":[143],"proposed":[145],"framework":[146],"submodular":[148],"maximization":[149],"characterize":[151],"its":[152],"connection":[153],"free":[156],"energy.":[157],"Experiments":[158],"variety":[161],"real-world":[163],"tasks":[164],"demonstrate":[165],"consistent":[166],"improvements":[167],"over":[168],"existing":[169],"baselines.":[170]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-15T00:00:00"}