{"id":"https://openalex.org/W7223069860","doi":"https://doi.org/10.48550/arxiv.2610.26285","title":"Learning SOCP Solver Dynamics with Graph Neural Networks","display_name":"Learning SOCP Solver Dynamics with Graph Neural Networks","publication_year":2026,"publication_date":"2026-10-19","ids":{"openalex":"https://openalex.org/W7223069860","doi":"https://doi.org/10.48550/arxiv.2610.26285"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.26285","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.26285","pdf_url":"https://arxiv.org/pdf/2610.26285","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/2610.26285","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5159105804","display_name":"Adam Ghalem","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I277688954","display_name":"Universit\u00e9 Paris-Saclay","ror":"https://ror.org/03xjwb503","country_code":"FR","type":"education","lineage":["https://openalex.org/I277688954"]},{"id":"https://openalex.org/I4210107720","display_name":"CentraleSup\u00e9lec","ror":"https://ror.org/019tcpt25","country_code":"FR","type":"facility","lineage":["https://openalex.org/I277688954","https://openalex.org/I4210107720"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Ghalem, Adam","raw_affiliation_strings":["Nokia Bell Labs","Universit\u00e9 Paris-Saclay , CentraleSup\u00e9lec , Inria"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nokia Bell Labs","institution_ids":[]},{"raw_affiliation_string":"Universit\u00e9 Paris-Saclay , CentraleSup\u00e9lec , Inria","institution_ids":["https://openalex.org/I277688954","https://openalex.org/I4210107720"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029552750","display_name":"Lou Sala\u00fcn","orcid":"https://orcid.org/0000-0003-0522-0435","observed_orcids":["https://orcid.org/0000-0003-0522-0435"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sala\u00fcn, Lou","raw_affiliation_strings":["Nokia Bell Labs"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nokia Bell Labs","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5146976895","display_name":"Davide Rinaldi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rinaldi, Davide","raw_affiliation_strings":["Nokia Bell Labs"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nokia Bell Labs","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5087251228","display_name":"Fragkiskos D. Malliaros","orcid":"https://orcid.org/0000-0002-8770-3969","observed_orcids":["https://orcid.org/0000-0002-8770-3969"]},"institutions":[{"id":"https://openalex.org/I277688954","display_name":"Universit\u00e9 Paris-Saclay","ror":"https://ror.org/03xjwb503","country_code":"FR","type":"education","lineage":["https://openalex.org/I277688954"]},{"id":"https://openalex.org/I4210107720","display_name":"CentraleSup\u00e9lec","ror":"https://ror.org/019tcpt25","country_code":"FR","type":"facility","lineage":["https://openalex.org/I277688954","https://openalex.org/I4210107720"]}],"countries":["FR"],"is_corresponding":false,"raw_author_name":"Malliaros, Fragkiskos D.","raw_affiliation_strings":["Universit\u00e9 Paris-Saclay , CentraleSup\u00e9lec , Inria"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universit\u00e9 Paris-Saclay , CentraleSup\u00e9lec , Inria","institution_ids":["https://openalex.org/I277688954","https://openalex.org/I4210107720"]}]}],"institutions":[],"countries_distinct_count":1,"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/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.98580002784729,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10963","display_name":"Advanced Optimization Algorithms Research","score":0.98580002784729,"subfield":{"id":"https://openalex.org/subfields/2612","display_name":"Numerical Analysis"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.004800000227987766,"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/T10791","display_name":"Advanced Control Systems Optimization","score":0.001500000013038516,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/second-order-cone-programming","display_name":"second-order cone programming","score":0.9750000238418579},{"id":"https://openalex.org/keywords/graph-neural-networks","display_name":"graph neural networks","score":0.9240000247955322},{"id":"https://openalex.org/keywords/recursive-neural-networks","display_name":"recursive neural networks","score":0.5910000205039978},{"id":"https://openalex.org/keywords/hybrid-optimization","display_name":"hybrid optimization","score":0.6880000233650208}],"concepts":[{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.7182999849319458},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6294999718666077},{"id":"https://openalex.org/C140479938","wikidata":"https://www.wikidata.org/wiki/Q5254619","display_name":"Iterated function","level":2,"score":0.5127999782562256},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.48969998955726624},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.4896000027656555},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.45239999890327454},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41119998693466187},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.39879998564720154},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.39809998869895935},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.36550000309944153},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.35929998755455017},{"id":"https://openalex.org/C157972887","wikidata":"https://www.wikidata.org/wiki/Q463359","display_name":"Convex optimization","level":3,"score":0.35850000381469727},{"id":"https://openalex.org/C81845259","wikidata":"https://www.wikidata.org/wiki/Q290117","display_name":"Quadratic programming","level":2,"score":0.3353999853134155},{"id":"https://openalex.org/C112680207","wikidata":"https://www.wikidata.org/wiki/Q714886","display_name":"Regular polygon","level":2,"score":0.33320000767707825},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.3221000134944916},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.31709998846054077},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C143587482","wikidata":"https://www.wikidata.org/wiki/Q1543216","display_name":"Iterative and incremental development","level":2,"score":0.3052999973297119},{"id":"https://openalex.org/C155253501","wikidata":"https://www.wikidata.org/wiki/Q461992","display_name":"Interior point method","level":2,"score":0.28360000252723694},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2782999873161316},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.27790001034736633},{"id":"https://openalex.org/C150452318","wikidata":"https://www.wikidata.org/wiki/Q4820432","display_name":"Augmented Lagrangian method","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C2779982483","wikidata":"https://www.wikidata.org/wiki/Q6094420","display_name":"Iterative refinement","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2621999979019165},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.25839999318122864},{"id":"https://openalex.org/C145446738","wikidata":"https://www.wikidata.org/wiki/Q319913","display_name":"Convex function","level":3,"score":0.25540000200271606}],"mesh":[],"locations_count":3,"locations":[{"id":"pmh:oai:arXiv.org:2610.26285","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.26285","pdf_url":"https://arxiv.org/pdf/2610.26285","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":"pmh:oai:HAL:hal-05804470v1","doi":null,"is_oa":true,"landing_page_url":"https://hal.science/hal-05804470","pdf_url":null,"source":{"id":"https://openalex.org/S4306402512","display_name":"HAL","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I4405252976","host_organization_name":"Centre pour la Communication Scientifique Directe","host_organization_lineage":["https://openalex.org/I4405252976"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Learning on Graphs Conference, Nov 2026, Boston, United States","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"doi:10.48550/arxiv.2610.26285","doi":"https://doi.org/10.48550/arxiv.2610.26285","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.26285","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":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.26285","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.26285","pdf_url":"https://arxiv.org/pdf/2610.26285","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":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7223069860.pdf","grobid_xml":"https://content.openalex.org/works/W7223069860.grobid-xml"},"referenced_works_count":3,"referenced_works":["https://openalex.org/W4323527626","https://openalex.org/W4404650153","https://openalex.org/W4407186839"],"related_works":["https://openalex.org/W2170161701","https://openalex.org/W4416635502","https://openalex.org/W7159660516","https://openalex.org/W2062903625","https://openalex.org/W7171656507","https://openalex.org/W2105235982","https://openalex.org/W4407186839","https://openalex.org/W4403663663","https://openalex.org/W2042440488","https://openalex.org/W2284305453"],"abstract_inverted_index":{"Second-order":[0],"cone":[1,62],"programs":[2],"(SOCPs)":[3],"are":[4,20],"a":[5,87,107,140,151,165,192,206,212],"fundamental":[6],"class":[7],"of":[8,143,221],"convex":[9],"optimization":[10,84,208],"problems":[11],"widely":[12],"applied":[13,127],"across":[14],"engineering":[15],"domains.":[16],"Classical":[17],"interior-point":[18,214],"methods":[19],"robust":[21],"and":[22,60,75,136,180],"accurate,":[23],"but":[24],"their":[25,65],"iterative":[26,118],"nature":[27],"can":[28,202],"be":[29,203],"costly":[30],"in":[31,150,205],"real-time":[32],"or":[33],"repeated-solve":[34],"settings.":[35],"This":[36],"has":[37,50],"motivated":[38],"learning-based":[39],"solvers":[40,81],"that":[41,111,182],"approximate":[42],"optimizer":[43,93],"solutions":[44],"with":[45],"neural":[46,54],"networks.":[47],"Recent":[48],"work":[49],"increasingly":[51],"studied":[52],"graph":[53],"networks":[55],"(GNNs)":[56],"for":[57,139],"linear,":[58],"quadratic,":[59],"second-order":[61],"optimization,":[63],"leveraging":[64],"ability":[66],"to":[67,115,159,191],"model":[68,125],"sparse":[69],"relational":[70],"structure":[71],"among":[72],"variables,":[73],"constraints,":[74],"objective":[76],"functions.":[77],"Most":[78],"existing":[79],"GNN-based":[80],"compress":[82],"the":[83,92,97,130,148,186],"process":[85],"into":[86],"single":[88,152],"forward":[89],"pass,":[90,153],"predicting":[91,147],"directly":[94],"without":[95,219],"using":[96,164],"intermediate":[98,157],"trajectory":[99],"produced":[100],"by":[101,210],"classical":[102,213],"solvers.":[103],"We":[104,169],"introduce":[105],"SOCPNet,":[106],"recursive":[108],"GNN":[109,124],"framework":[110],"uses":[112],"solver":[113,217],"trajectories":[114],"learn":[116],"an":[117],"SOCP":[119,177],"solution":[120,149,188],"procedure.":[121,168],"A":[122],"shared":[123],"is":[126],"repeatedly,":[128],"refining":[129],"current":[131],"iterate":[132],"at":[133],"each":[134],"step":[135],"enabling":[137],"inference":[138],"variable":[141],"number":[142],"iterations.":[144],"Rather":[145],"than":[146],"SOCPNet":[154,171,200],"learns":[155],"from":[156],"iterates":[158],"progressively":[160],"refine":[161],"its":[162],"estimates":[163],"trajectory-regularized":[166],"training":[167],"evaluate":[170],"on":[172],"multiple":[173],"datasets":[174],"spanning":[175],"standard":[176],"problem":[178],"classes":[179],"show":[181],"it":[183],"consistently":[184],"reduces":[185],"final":[187],"error":[189],"relative":[190],"SOCP-GNN":[193],"baseline.":[194],"Finally,":[195],"we":[196],"demonstrate":[197],"how":[198],"trained":[199],"models":[201],"used":[204],"hybrid":[207],"pipeline":[209],"warm-starting":[211],"solver,":[215],"reducing":[216],"iterations":[218],"loss":[220],"accuracy.":[222]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-21T00:00:00"}