{"id":"https://openalex.org/W7159043698","doi":"https://doi.org/10.48550/arxiv.2604.26470","title":"Hierarchical adaptive control for real-time dynamic inference at the edge","display_name":"Hierarchical adaptive control for real-time dynamic inference at the edge","publication_year":2026,"publication_date":"2026-04-29","ids":{"openalex":"https://openalex.org/W7159043698","doi":"https://doi.org/10.48550/arxiv.2604.26470"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.26470","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.26470","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2604.26470","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5019082004","display_name":"Francesco Daghero","orcid":"https://orcid.org/0000-0001-9595-7216"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Daghero, Francesco","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018940565","display_name":"Mahyar Tourchi Moghaddam","orcid":"https://orcid.org/0000-0001-5028-7546"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Moghaddam, Mahyar Tourchi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129764523","display_name":"Mikkel Baun Kj\u00e6rgaard","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kj\u00e6rgaard, Mikkel Baun","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/T10933","display_name":"Real-Time Systems Scheduling","score":0.2912999987602234,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10933","display_name":"Real-Time Systems Scheduling","score":0.2912999987602234,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.07729999721050262,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T12761","display_name":"Data Stream Mining Techniques","score":0.07460000365972519,"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/latency","display_name":"Latency (audio)","score":0.7418000102043152},{"id":"https://openalex.org/keywords/efficient-energy-use","display_name":"Efficient energy use","score":0.5209000110626221},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.4262000024318695},{"id":"https://openalex.org/keywords/controller","display_name":"Controller (irrigation)","score":0.40950000286102295},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.40540000796318054},{"id":"https://openalex.org/keywords/dram","display_name":"Dram","score":0.36629998683929443},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.357699990272522}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7544999718666077},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.7418000102043152},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.5209000110626221},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.4844000041484833},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.4799000024795532},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.4262000024318695},{"id":"https://openalex.org/C203479927","wikidata":"https://www.wikidata.org/wiki/Q5165939","display_name":"Controller (irrigation)","level":2,"score":0.40950000286102295},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.40540000796318054},{"id":"https://openalex.org/C7366592","wikidata":"https://www.wikidata.org/wiki/Q1255620","display_name":"Dram","level":2,"score":0.36629998683929443},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.357699990272522},{"id":"https://openalex.org/C2778456923","wikidata":"https://www.wikidata.org/wiki/Q5337692","display_name":"Edge computing","level":3,"score":0.3463999927043915},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.3118000030517578},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.30640000104904175},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.2840000092983246},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C189930140","wikidata":"https://www.wikidata.org/wiki/Q1112878","display_name":"CAS latency","level":4,"score":0.26019999384880066},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C19012869","wikidata":"https://www.wikidata.org/wiki/Q578372","display_name":"Response time","level":2,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.26470","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.26470","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.26470","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.26470","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.6515989303588867,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Industrial":[0],"systems":[1],"increasingly":[2],"depend":[3],"on":[4,10,69,199,259],"Machine":[5],"Learning":[6],"(ML),":[7],"and":[8,19,36,106,116,131,137,150,164,183,216,251,254],"operate":[9],"heterogeneous":[11],"nodes":[12],"that":[13,73,82,103,238,246],"must":[14,74],"satisfy":[15],"tight":[16],"latency,":[17],"energy,":[18],"memory":[20,138],"constraints.":[21,139],"Dynamic":[22],"ML":[23],"models,":[24],"which":[25],"reconfigure":[26],"their":[27,45],"computational":[28],"footprint":[29],"at":[30],"runtime,":[31],"promise":[32],"high":[33,161],"energy":[34,162,217],"efficiency":[35,163,248],"lower":[37],"average":[38,63,208],"latency":[39,64,136,212,241],"for":[40,118],"modest":[41],"accuracy":[42,61,225],"tradeoffs;":[43],"however,":[44],"deployment":[46],"is":[47],"complex":[48],"due":[49],"to":[50,89,159,214,219,228],"the":[51,60,76,110,141,197],"additional":[52],"hyperparameters":[53],"they":[54],"rely":[55],"on.":[56],"These":[57],"hyperparameters,":[58],"controlling":[59],"versus":[62],"tradeoff,":[65],"are":[66],"often":[67],"tuned":[68],"a":[70,99,113,122,132,144,179,234,243],"calibration":[71],"dataset":[72],"match":[75],"test":[77],"time":[78],"distribution,":[79],"an":[80,190,256],"assumption":[81],"rarely":[83],"holds":[84],"in":[85,187],"real-world":[86],"scenarios,":[87,206],"leading":[88],"suboptimal":[90],"operational":[91],"conditions,":[92],"possibly":[93],"below":[94],"static":[95,229],"models.":[96],"We":[97,195],"propose":[98],"two-tier":[100],"adaptive":[101],"architecture":[102],"co-optimizes":[104],"model":[105],"system":[107],"decisions.":[108],"At":[109,140],"global":[111,180,193],"level,":[112,143],"scheduler":[114],"configures":[115],"deploys,":[117],"each":[119],"edge":[120],"node,":[121],"cascade":[123],"of":[124,127,189,211],"classifiers":[125],"composed":[126],"lightweight":[128],"specialized":[129,156],"models":[130],"generalist":[133],"fallback,":[134],"satisfying":[135],"node":[142],"local":[145],"controller":[146,245],"tracks":[147],"data":[148,250],"drifts":[149],"hardware":[151],"resources,":[152],"enabling":[153],"or":[154],"disabling":[155],"predictors":[157],"(SP)":[158],"preserve":[160],"avoid":[165],"latency-constraint":[166],"violations":[167],"under":[168,202,249],"varying":[169],"conditions.":[170],"This":[171],"design":[172],"allows":[173],"longer":[174],"operating":[175],"times":[176],"without":[177],"forcing":[178],"redeployment":[181],"step,":[182],"enables":[184],"efficient":[185],"execution":[186],"case":[188],"unreachable":[191],"remote":[192],"controller.":[194],"evaluate":[196],"approach":[198],"two":[200],"datasets":[201],"controlled":[203],"distribution":[204],"mismatch":[205],"showing":[207],"per-inference":[209],"reductions":[210],"up":[213,218],"2.45x":[215],"2.86x,":[220],"with":[221],"less":[222],"than":[223],"4%":[224],"drop":[226],"compared":[227],"baselines.":[230],"Our":[231],"contributions":[232],"are:(1)":[233],"budgeted":[235],"SP-cascade":[236],"formulation":[237],"preserves":[239],"worst-case":[240],"constraints;(2)":[242],"hierarchical":[244],"maintains":[247],"resource":[252],"changes;":[253],"(3)":[255],"experimental":[257],"evaluation":[258],"embedded":[260],"hardware.":[261]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-01T00:00:00"}