{"id":"https://openalex.org/W4414580406","doi":"https://doi.org/10.48550/arxiv.2502.19377","title":"Preference-Based Gradient Estimation for ML-Guided Approximate Combinatorial Optimization","display_name":"Preference-Based Gradient Estimation for ML-Guided Approximate Combinatorial Optimization","publication_year":2025,"publication_date":"2025-02-26","ids":{"openalex":"https://openalex.org/W4414580406","doi":"https://doi.org/10.48550/arxiv.2502.19377"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2502.19377","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.19377","pdf_url":"https://arxiv.org/pdf/2502.19377","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/2502.19377","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Mielke, Arman","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mielke, Arman","raw_affiliation_strings":["ETAS Research Stuttgart , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ETAS Research Stuttgart , Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089873883","display_name":"Uwe Bauknecht","orcid":"https://orcid.org/0009-0003-3164-1812","observed_orcids":["https://orcid.org/0009-0003-3164-1812"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bauknecht, Uwe","raw_affiliation_strings":["ETAS Research Stuttgart , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ETAS Research Stuttgart , Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078058219","display_name":"Thilo Strauss","orcid":"https://orcid.org/0000-0002-6712-2171","observed_orcids":["https://orcid.org/0000-0002-6712-2171"]},"institutions":[{"id":"https://openalex.org/I69356397","display_name":"Xi\u2019an Jiaotong-Liverpool University","ror":"https://ror.org/03zmrmn05","country_code":"CN","type":"education","lineage":["https://openalex.org/I69356397"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Strauss, Thilo","raw_affiliation_strings":["School of AI and Advanced Computing Xi' an Jiaotong-Liverpool University China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of AI and Advanced Computing Xi' an Jiaotong-Liverpool University China","institution_ids":["https://openalex.org/I69356397"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5031719069","display_name":"Mathias Niepert","orcid":"https://orcid.org/0000-0002-8401-3751","observed_orcids":["https://orcid.org/0000-0002-8401-3751"]},"institutions":[{"id":"https://openalex.org/I100066346","display_name":"University of Stuttgart","ror":"https://ror.org/04vnq7t77","country_code":"DE","type":"education","lineage":["https://openalex.org/I100066346"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Niepert, Mathias","raw_affiliation_strings":["Computer Science Department University of Stuttgart Stuttgart , Germany"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computer Science Department University of Stuttgart Stuttgart , Germany","institution_ids":["https://openalex.org/I100066346"]}]}],"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":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/T10720","display_name":"Complexity and Algorithms in Graphs","score":0.7832000255584717,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10720","display_name":"Complexity and Algorithms in Graphs","score":0.7832000255584717,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10100","display_name":"Metaheuristic Optimization Algorithms Research","score":0.1185000017285347,"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":0.019500000402331352,"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/combinatorial-optimization","display_name":"combinatorial optimization","score":0.9559999704360962},{"id":"https://openalex.org/keywords/approximate-optimization","display_name":"approximate optimization","score":0.5389999747276306},{"id":"https://openalex.org/keywords/graph-neural-networks","display_name":"graph neural networks","score":0.8220000267028809},{"id":"https://openalex.org/keywords/gradient-estimation","display_name":"gradient estimation","score":0.7699999809265137},{"id":"https://openalex.org/keywords/preference-learning","display_name":"preference learning","score":0.6100000143051147},{"id":"https://openalex.org/keywords/self-supervised-learning","display_name":"self-supervised learning","score":0.8050000071525574},{"id":"https://openalex.org/keywords/traveling-salesman-problem","display_name":"traveling salesman problem","score":0.765999972820282}],"concepts":[{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.6087999939918518},{"id":"https://openalex.org/C148764684","wikidata":"https://www.wikidata.org/wiki/Q621751","display_name":"Approximation algorithm","level":2,"score":0.552299976348877},{"id":"https://openalex.org/C175859090","wikidata":"https://www.wikidata.org/wiki/Q322212","display_name":"Travelling salesman problem","level":2,"score":0.5415999889373779},{"id":"https://openalex.org/C52692508","wikidata":"https://www.wikidata.org/wiki/Q1333872","display_name":"Combinatorial optimization","level":2,"score":0.5189999938011169},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47600001096725464},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.4693000018596649},{"id":"https://openalex.org/C188919014","wikidata":"https://www.wikidata.org/wiki/Q5422296","display_name":"Extremal optimization","level":4,"score":0.45660001039505005},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.43650001287460327},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4000000059604645},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.39480000734329224},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.3734000027179718},{"id":"https://openalex.org/C75782508","wikidata":"https://www.wikidata.org/wiki/Q3333633","display_name":"Cross-entropy method","level":4,"score":0.3555000126361847},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.32839998602867126},{"id":"https://openalex.org/C156778621","wikidata":"https://www.wikidata.org/wiki/Q1365748","display_name":"Spectrum (functional analysis)","level":2,"score":0.3199999928474426},{"id":"https://openalex.org/C106472803","wikidata":"https://www.wikidata.org/wiki/Q291440","display_name":"2-opt","level":3,"score":0.30880001187324524},{"id":"https://openalex.org/C2987595161","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Optimization algorithm","level":2,"score":0.29980000853538513},{"id":"https://openalex.org/C145242015","wikidata":"https://www.wikidata.org/wiki/Q774123","display_name":"Approximation theory","level":2,"score":0.2962000072002411},{"id":"https://openalex.org/C55479107","wikidata":"https://www.wikidata.org/wiki/Q97663916","display_name":"Stochastic approximation","level":3,"score":0.2791000008583069},{"id":"https://openalex.org/C115680565","wikidata":"https://www.wikidata.org/wiki/Q5977448","display_name":"Gradient method","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.272599995136261},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2639999985694885},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2630999982357025},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.2581000030040741},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.2574999928474426},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.25690001249313354},{"id":"https://openalex.org/C99140742","wikidata":"https://www.wikidata.org/wiki/Q843550","display_name":"Polynomial-time approximation scheme","level":3,"score":0.25029999017715454}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2502.19377","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.19377","pdf_url":"https://arxiv.org/pdf/2502.19377","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.2502.19377","doi":"https://doi.org/10.48550/arxiv.2502.19377","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2502.19377","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:2502.19377","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2502.19377","pdf_url":"https://arxiv.org/pdf/2502.19377","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":[{"id":"https://openalex.org/F4320317967","display_name":"International Max Planck Research School for Environmental, Cellular and Molecular Microbiology","ror":null},{"id":"https://openalex.org/F4320327553","display_name":"International Max Planck Research School for Advanced Methods in Process and Systems Engineering","ror":"https://ror.org/04hazks31"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4414580406.pdf","grobid_xml":"https://content.openalex.org/works/W4414580406.grobid-xml"},"referenced_works_count":10,"referenced_works":["https://openalex.org/W2009339416","https://openalex.org/W2056898464","https://openalex.org/W2119717200","https://openalex.org/W2768242641","https://openalex.org/W2948433391","https://openalex.org/W3010642661","https://openalex.org/W4288055665","https://openalex.org/W4382318472","https://openalex.org/W4394657357","https://openalex.org/W4400222632"],"related_works":["https://openalex.org/W2607264901","https://openalex.org/W4399453936","https://openalex.org/W4225733855","https://openalex.org/W4319451700","https://openalex.org/W2560592986","https://openalex.org/W4390833107","https://openalex.org/W7106294934","https://openalex.org/W4378760003","https://openalex.org/W7133235363","https://openalex.org/W3213494343"],"abstract_inverted_index":{"Combinatorial":[0],"optimization":[1,130],"(CO)":[2],"problems":[3],"arise":[4],"across":[5],"a":[6,30,39,76,80,91],"broad":[7],"spectrum":[8],"of":[9,99],"domains,":[10],"including":[11],"medicine,":[12],"logistics,":[13],"and":[14,56,101,137],"manufacturing.":[15],"While":[16],"exact":[17],"solutions":[18,28],"are":[19],"often":[20],"computationally":[21],"infeasible,":[22],"many":[23],"practical":[24],"applications":[25],"require":[26],"high-quality":[27],"within":[29],"given":[31],"time":[32],"budget.":[33],"To":[34],"address":[35],"this,":[36],"we":[37,51],"propose":[38],"learning-based":[40],"approach":[41,95,148],"that":[42,66,85,145],"enhances":[43],"existing":[44],"non-learned":[45],"approximation":[46,54,88,118],"algorithms":[47,55],"for":[48],"CO.":[49],"Specifically,":[50],"parameterize":[52],"these":[53],"train":[57],"graph":[58],"neural":[59],"networks":[60],"(GNNs)":[61],"to":[62,109],"predict":[63],"parameter":[64],"values":[65],"yield":[67],"near-optimal":[68],"solutions.":[69],"Our":[70,142],"method":[71,125],"is":[72,149],"trained":[73],"end-to-end":[74],"in":[75],"self-supervised":[77],"fashion,":[78],"using":[79],"novel":[81],"gradient":[82],"estimation":[83],"scheme":[84],"treats":[86],"the":[87,97,104,111,117,132,138,146],"algorithm":[89,112,119],"as":[90],"black":[92],"box.":[93],"This":[94],"combines":[96],"strengths":[98],"learning":[100],"traditional":[102],"algorithms:":[103],"GNN":[105],"learns":[106],"from":[107],"data":[108],"guide":[110],"toward":[113],"better":[114],"solutions,":[115],"while":[116],"ensures":[120],"feasibility.":[121],"We":[122],"validate":[123],"our":[124],"on":[126],"two":[127],"well-known":[128],"combinatorial":[129],"problems:":[131],"travelling":[133],"salesman":[134],"problem":[135],"(TSP)":[136],"minimum":[139],"k-cut":[140],"problem.":[141],"results":[143],"demonstrate":[144],"proposed":[147],"competitive":[150],"with":[151],"state-of-the-art":[152],"learned":[153],"CO":[154],"solvers.":[155]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}