{"id":"https://openalex.org/W4414433432","doi":"https://doi.org/10.1109/case58245.2025.11163833","title":"OffLight: An Offline Multi-Agent Reinforcement Learning Framework for Traffic Signal Control","display_name":"OffLight: An Offline Multi-Agent Reinforcement Learning Framework for Traffic Signal Control","publication_year":2025,"publication_date":"2025-08-17","ids":{"openalex":"https://openalex.org/W4414433432","doi":"https://doi.org/10.1109/case58245.2025.11163833"},"language":"en","primary_location":{"id":"doi:10.1109/case58245.2025.11163833","is_oa":false,"landing_page_url":"https://doi.org/10.1109/case58245.2025.11163833","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 21st International Conference on Automation Science and Engineering (CASE)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102730372","display_name":"Rohit Bokade","orcid":"https://orcid.org/0009-0005-5893-0969"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rohit Bokade","raw_affiliation_strings":["Northeastern University,Department of Mechanical and Industrial Engineering,Boston,MA,USA,02115"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University,Department of Mechanical and Industrial Engineering,Boston,MA,USA,02115","institution_ids":["https://openalex.org/I12912129"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5050575763","display_name":"Xiaoning Jin","orcid":"https://orcid.org/0000-0001-9353-8456"},"institutions":[{"id":"https://openalex.org/I12912129","display_name":"Northeastern University","ror":"https://ror.org/04t5xt781","country_code":"US","type":"education","lineage":["https://openalex.org/I12912129"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaoning Jin","raw_affiliation_strings":["Northeastern University,Department of Mechanical and Industrial Engineering,Boston,MA,USA,02115"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Northeastern University,Department of Mechanical and Industrial Engineering,Boston,MA,USA,02115","institution_ids":["https://openalex.org/I12912129"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I12912129"],"apc_list":null,"apc_paid":null,"fwci":3.3874,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.91294724,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"2730","last_page":"2737"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10524","display_name":"Traffic control and management","score":0.9883999824523926,"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"}},"topics":[{"id":"https://openalex.org/T10524","display_name":"Traffic control and management","score":0.9883999824523926,"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"}},{"id":"https://openalex.org/T11344","display_name":"Traffic Prediction and Management Techniques","score":0.9248999953269958,"subfield":{"id":"https://openalex.org/subfields/2215","display_name":"Building and Construction"},"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":0.754800021648407},{"id":"https://openalex.org/keywords/autoencoder","display_name":"Autoencoder","score":0.5309000015258789},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5217999815940857},{"id":"https://openalex.org/keywords/online-and-offline","display_name":"Online and offline","score":0.4171999990940094},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.35519999265670776},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.3499999940395355},{"id":"https://openalex.org/keywords/suite","display_name":"Suite","score":0.34689998626708984},{"id":"https://openalex.org/keywords/traffic-congestion","display_name":"Traffic congestion","score":0.34630000591278076},{"id":"https://openalex.org/keywords/traffic-generation-model","display_name":"Traffic generation model","score":0.33149999380111694}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.754800021648407},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7530999779701233},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.5309000015258789},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5217999815940857},{"id":"https://openalex.org/C2780102126","wikidata":"https://www.wikidata.org/wiki/Q10928179","display_name":"Online and offline","level":2,"score":0.4171999990940094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.413100004196167},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.40450000762939453},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3700000047683716},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.35519999265670776},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3499999940395355},{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.34689998626708984},{"id":"https://openalex.org/C2779888511","wikidata":"https://www.wikidata.org/wiki/Q244156","display_name":"Traffic congestion","level":2,"score":0.34630000591278076},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3424000144004822},{"id":"https://openalex.org/C176715033","wikidata":"https://www.wikidata.org/wiki/Q2080768","display_name":"Traffic generation model","level":2,"score":0.33149999380111694},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.33090001344680786},{"id":"https://openalex.org/C2778391309","wikidata":"https://www.wikidata.org/wiki/Q7832527","display_name":"Traffic simulation","level":3,"score":0.32899999618530273},{"id":"https://openalex.org/C2780490138","wikidata":"https://www.wikidata.org/wiki/Q7079636","display_name":"Offline learning","level":3,"score":0.3197000026702881},{"id":"https://openalex.org/C22684755","wikidata":"https://www.wikidata.org/wiki/Q847526","display_name":"Queueing theory","level":2,"score":0.30660000443458557},{"id":"https://openalex.org/C2781395549","wikidata":"https://www.wikidata.org/wiki/Q4680762","display_name":"Adaptive sampling","level":3,"score":0.3027999997138977},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.30239999294281006},{"id":"https://openalex.org/C107464732","wikidata":"https://www.wikidata.org/wiki/Q235781","display_name":"Adaptive control","level":3,"score":0.29899999499320984},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.29899999499320984},{"id":"https://openalex.org/C160403385","wikidata":"https://www.wikidata.org/wiki/Q220543","display_name":"Queue","level":2,"score":0.29809999465942383},{"id":"https://openalex.org/C73602740","wikidata":"https://www.wikidata.org/wiki/Q7795822","display_name":"Thompson sampling","level":3,"score":0.2782000005245209},{"id":"https://openalex.org/C2777113093","wikidata":"https://www.wikidata.org/wiki/Q221488","display_name":"Pedestrian","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.26579999923706055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/case58245.2025.11163833","is_oa":false,"landing_page_url":"https://doi.org/10.1109/case58245.2025.11163833","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE 21st International Conference on Automation Science and Engineering (CASE)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":14,"referenced_works":["https://openalex.org/W192920577","https://openalex.org/W2145339207","https://openalex.org/W2617547828","https://openalex.org/W2747213132","https://openalex.org/W2964749398","https://openalex.org/W2983178256","https://openalex.org/W2985159211","https://openalex.org/W3044015199","https://openalex.org/W3095594668","https://openalex.org/W4220899555","https://openalex.org/W4384202122","https://openalex.org/W4400147025","https://openalex.org/W4401863796","https://openalex.org/W4404576982"],"related_works":[],"abstract_inverted_index":{"Efficient":[0],"traffic":[1,24,74,117,129,136,198],"signal":[2,75,137],"control":[3,76],"(TSC)":[4],"is":[5,204],"crucial":[6],"for":[7,85,183],"reducing":[8,149],"congestion":[9],"and":[10,35,47,88,115,119,157,176],"improving":[11],"urban":[12,128,197],"mobility.":[13],"While":[14],"multi-agent":[15],"reinforcement":[16],"learning":[17],"(MARL)":[18],"has":[19],"shown":[20],"promise":[21],"in":[22,57,73],"adaptive":[23],"management,":[25],"its":[26],"reliance":[27],"on":[28],"real-time":[29],"interactions":[30],"makes":[31],"online":[32,188],"training":[33],"costly":[34],"impractical.":[36],"Offline":[37],"MARL,":[38],"which":[39,112],"learns":[40],"from":[41,132],"historical":[42],"data,":[43],"offers":[44],"a":[45,63,104],"safer":[46],"more":[48,181],"scalable":[49],"alternative":[50],"but":[51],"struggles":[52],"with":[53],"heterogeneous":[54],"behavior":[55,100],"policies":[56],"real-world":[58,127,184],"datasets.":[59,77],"We":[60,123],"introduce":[61],"OffLight,":[62],"novel":[64],"offline":[65,145,168],"MARL":[66],"framework":[67],"designed":[68],"to":[69,83,93,134,155,163,174,195],"handle":[70],"policy":[71,121],"heterogeneity":[72],"OffLight":[78,125,142,171],"integrates":[79],"importance":[80],"sampling":[81,91],"(IS)":[82],"correct":[84],"distributional":[86],"shifts":[87],"return-based":[89],"prioritized":[90],"(RBPS)":[92],"emphasize":[94],"high-quality":[95],"experiences.":[96],"To":[97],"model":[98,107],"diverse":[99],"policies,":[101],"it":[102,180],"leverages":[103],"Gaussian":[105],"mixture":[106],"variational":[108],"graph":[109],"autoencoder":[110],"(GMM-VGAE)":[111],"captures":[113],"spatial":[114],"temporal":[116],"dynamics":[118],"improves":[120],"estimation.":[122],"evaluate":[124],"across":[126],"scenarios":[130],"varying":[131],"small":[133],"large":[135],"networks.":[138],"Results":[139],"show":[140],"that":[141],"outperforms":[143],"existing":[144],"RL":[146,169],"methods":[147],"by":[148,153,160],"average":[150],"travel":[151],"time":[152],"up":[154],"7.8%":[156],"queue":[158],"length":[159],"11.2%":[161],"compared":[162],"state-of-the-art":[164],"approaches.":[165],"Unlike":[166],"prior":[167],"methods,":[170],"effectively":[172],"adapts":[173],"complex":[175],"mixed-policy":[177],"datasets,":[178],"making":[179],"reliable":[182],"deployment":[185],"without":[186],"risky":[187],"training.":[189],"These":[190],"results":[191],"highlight":[192],"OffLight\u2019s":[193],"potential":[194],"improve":[196],"management":[199],"at":[200,206],"scale.":[201],"Our":[202],"implementation":[203],"available":[205],"https://github.com/rbokade/offlight.":[207]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}