{"id":"https://openalex.org/W7135081192","doi":"https://doi.org/10.48550/arxiv.2603.10053","title":"Cluster-Aware Attention-Based Deep Reinforcement Learning for Pickup and Delivery Problems","display_name":"Cluster-Aware Attention-Based Deep Reinforcement Learning for Pickup and Delivery Problems","publication_year":2026,"publication_date":"2026-03-09","ids":{"openalex":"https://openalex.org/W7135081192","doi":"https://doi.org/10.48550/arxiv.2603.10053"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.10053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10053","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.10053","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128878189","display_name":"Wentao Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Wentao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128888425","display_name":"Lifeng Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Lifeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5083864266","display_name":"Guangyu Zou","orcid":"https://orcid.org/0000-0003-1890-6874"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zou, Guangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T10567","display_name":"Vehicle Routing Optimization Methods","score":0.9620000123977661,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10567","display_name":"Vehicle Routing Optimization Methods","score":0.9620000123977661,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11942","display_name":"Transportation and Mobility Innovations","score":0.005100000184029341,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10761","display_name":"Vehicular Ad Hoc Networks (VANETs)","score":0.002899999963119626,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic 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/reinforcement-learning","display_name":"Reinforcement learning","score":0.8093000054359436},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5386999845504761},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4781000018119812},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4740000069141388},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.4618000090122223},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.41769999265670776},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.3928000032901764}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.8093000054359436},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.76910001039505},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5386999845504761},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4781000018119812},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4740000069141388},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.4618000090122223},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41920000314712524},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.41769999265670776},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3928000032901764},{"id":"https://openalex.org/C113954288","wikidata":"https://www.wikidata.org/wiki/Q186885","display_name":"Timestamp","level":2,"score":0.38670000433921814},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3767000138759613},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.35989999771118164},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.34529998898506165},{"id":"https://openalex.org/C2776221269","wikidata":"https://www.wikidata.org/wiki/Q572648","display_name":"Pickup","level":3,"score":0.32190001010894775},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.31209999322891235},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.29420000314712524},{"id":"https://openalex.org/C89109886","wikidata":"https://www.wikidata.org/wiki/Q1535924","display_name":"Trust region","level":3,"score":0.28760001063346863}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.10053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10053","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.10053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.10053","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"The":[0,96,148],"Pickup":[1],"and":[2,9,25,93,102,112,122,142,169,186,222],"Delivery":[3],"Problem":[4],"(PDP)":[5],"is":[6,150],"a":[7,43,78,100,131,135,154],"fundamental":[8],"challenging":[10],"variant":[11],"of":[12,64,87],"the":[13,62,84],"Vehicle":[14],"Routing":[15],"Problem,":[16],"characterized":[17],"by":[18],"tightly":[19],"coupled":[20],"pickup--delivery":[21],"pairs,":[22],"precedence":[23],"constraints,":[24,52],"spatial":[26],"layouts":[27],"that":[28,81,117,138,174,214],"often":[29],"exhibit":[30],"clustering.":[31],"Existing":[32],"deep":[33],"reinforcement":[34],"learning":[35,49],"(DRL)":[36],"approaches":[37],"either":[38],"model":[39],"all":[40],"nodes":[41],"on":[42,47,99,126,166,183,190],"flat":[44],"graph,":[45],"relying":[46],"implicit":[48],"to":[50],"enforce":[51],"or":[53,177],"achieve":[54],"strong":[55,180],"performance":[56],"through":[57],"inference-time":[58],"collaborative":[59],"search":[60],"at":[61,145],"cost":[63],"substantial":[65],"latency.":[66],"In":[67],"this":[68],"paper,":[69],"we":[70,129],"propose":[71],"\\emph{CAADRL}":[72],"(Cluster-Aware":[73],"Attention-based":[74],"Deep":[75],"Reinforcement":[76],"Learning),":[77],"DRL":[79],"framework":[80],"explicitly":[82,215],"exploits":[83],"multi-scale":[85],"structure":[86,218],"PDP":[88,171,228],"instances":[89,185],"via":[90],"cluster-aware":[91],"encoding":[92],"hierarchical":[94],"decoding.":[95],"encoder":[97],"builds":[98],"Transformer":[101],"combines":[103],"global":[104],"self-attention":[105],"with":[106,134,153,204],"intra-cluster":[107,140],"attention":[108],"over":[109],"depot,":[110],"pickup,":[111],"delivery":[113],"nodes,":[114],"producing":[115],"embeddings":[116],"are":[118],"both":[119],"globally":[120],"informative":[121],"locally":[123],"role-aware.":[124],"Based":[125],"these":[127,202],"embeddings,":[128],"introduce":[130],"Dynamic":[132],"Dual-Decoder":[133],"learnable":[136],"gate":[137],"balances":[139],"routing":[141],"inter-cluster":[143],"transitions":[144],"each":[146],"step.":[147],"policy":[149,156],"trained":[151],"end-to-end":[152],"POMO-style":[155],"gradient":[157],"scheme":[158],"using":[159],"multiple":[160],"symmetric":[161],"rollouts":[162],"per":[163],"instance.":[164],"Experiments":[165],"synthetic":[167],"clustered":[168,184],"uniform":[170,191],"benchmarks":[172],"show":[173],"CAADRL":[175],"matches":[176],"improves":[178],"upon":[179],"state-of-the-art":[181],"baselines":[182],"remains":[187],"highly":[188],"competitive":[189],"instances,":[192],"particularly":[193],"as":[194],"problem":[195],"size":[196],"increases.":[197],"Crucially,":[198],"our":[199],"method":[200],"achieves":[201],"results":[203],"substantially":[205],"lower":[206],"inference":[207],"time":[208],"than":[209],"neural":[210,227],"collaborative-search":[211],"baselines,":[212],"suggesting":[213],"modeling":[216],"cluster":[217],"provides":[219],"an":[220],"effective":[221],"efficient":[223],"inductive":[224],"bias":[225],"for":[226],"solvers.":[229]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-03-13T00:00:00"}