{"id":"https://openalex.org/W7204127251","doi":"https://doi.org/10.7840/kics.2026.51.8.1645","title":"A Safety-Preserving Inference Scheduling Framework for Physical AI under Non-stationary Energy Harvesting","display_name":"A Safety-Preserving Inference Scheduling Framework for Physical AI under Non-stationary Energy Harvesting","publication_year":2026,"publication_date":"2026-08-24","ids":{"openalex":"https://openalex.org/W7204127251","doi":"https://doi.org/10.7840/kics.2026.51.8.1645"},"language":"en","primary_location":{"id":"doi:10.7840/kics.2026.51.8.1645","doi":"https://doi.org/10.7840/kics.2026.51.8.1645","is_oa":false,"landing_page_url":"https://doi.org/10.7840/kics.2026.51.8.1645","pdf_url":null,"source":{"id":"https://openalex.org/S4306496539","display_name":"\ud55c\uad6d\ud1b5\uc2e0\ud559\ud68c\ub17c\ubb38\uc9c0","issn_l":"1226-4717","issn":["1226-4717","2287-3880"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["kci-registered"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Korean Institute of Communications and Information Sciences","raw_type":"journal-article"},"volume":null,"type":"article","is_traditional":true,"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/A5148007517","display_name":"Hye-min Shin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hye-min Shin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5059178827","display_name":"Young-myoung Kang","orcid":"https://orcid.org/0000-0001-8226-6440","observed_orcids":["https://orcid.org/0000-0001-8226-6440"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Young-myoung Kang","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"51","issue":"8","first_page":"1645","last_page":"1655"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10273","display_name":"IoT and Edge/Fog Computing","score":0.7753999829292297,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.7753999829292297,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11392","display_name":"Energy Harvesting in Wireless Networks","score":0.13269999623298645,"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"}},{"id":"https://openalex.org/T13553","display_name":"Age of Information Optimization","score":0.027899999171495438,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/physical-ai","display_name":"physical AI","score":0.8299999833106995},{"id":"https://openalex.org/keywords/inference-scheduling","display_name":"inference scheduling","score":0.8690000176429749},{"id":"https://openalex.org/keywords/energy-harvesting","display_name":"energy harvesting","score":0.8569999933242798},{"id":"https://openalex.org/keywords/battery-safety","display_name":"battery safety","score":0.824999988079071},{"id":"https://openalex.org/keywords/pareto-front","display_name":"Pareto front","score":0.9509999752044678},{"id":"https://openalex.org/keywords/resource-constrained-devices","display_name":"resource-constrained devices","score":0.7229999899864197}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6177999973297119},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.5342000126838684},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.39879998564720154},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36079999804496765},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.3384999930858612},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.32199999690055847},{"id":"https://openalex.org/C101518730","wikidata":"https://www.wikidata.org/wiki/Q930236","display_name":"Energy harvesting","level":3,"score":0.31439998745918274},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.266400009393692},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.25600001215934753},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.25130000710487366},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.25049999356269836}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.7840/kics.2026.51.8.1645","doi":"https://doi.org/10.7840/kics.2026.51.8.1645","is_oa":false,"landing_page_url":"https://doi.org/10.7840/kics.2026.51.8.1645","pdf_url":null,"source":{"id":"https://openalex.org/S4306496539","display_name":"\ud55c\uad6d\ud1b5\uc2e0\ud559\ud68c\ub17c\ubb38\uc9c0","issn_l":"1226-4717","issn":["1226-4717","2287-3880"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["kci-registered"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"The Journal of Korean Institute of Communications and Information Sciences","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.75,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"},{"score":0.5378999710083008,"id":"https://metadata.un.org/sdg/9","display_name":"Industry, innovation and infrastructure"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7172108552","https://openalex.org/W2982874467","https://openalex.org/W7201995704","https://openalex.org/W7138908638","https://openalex.org/W7165746464","https://openalex.org/W4285132705","https://openalex.org/W4386249473","https://openalex.org/W3166214794","https://openalex.org/W7171847834","https://openalex.org/W7161015704"],"abstract_inverted_index":{"Compact":[0],"Physical":[1,38,46,193],"AI":[2,39,47,194],"models":[3],"are":[4],"increasingly":[5],"deployed":[6],"on":[7],"energy-harvesting":[8],"(EH)":[9],"IoT":[10],"robots.":[11],"However,":[12],"existing":[13],"EH":[14,85,93,137,152,177],"schedulers":[15,79],"assume":[16],"either":[17],"binary":[18],"task":[19],"quality":[20,33,56,103,183],"or":[21,163],"a":[22,52,71,126],"continuous":[23],"quality-energy":[24],"trade-off,":[25],"neither":[26],"of":[27,75,169],"which":[28,115],"captures":[29],"the":[30,45,66,87,97,101,165,170],"discrete,":[31],"model-class-dependent":[32],"gaps":[34],"that":[35,78,92,155],"arise":[36],"in":[37],"tier":[40,133],"selection.":[41],"This":[42],"paper":[43],"introduces":[44],"Inference":[48],"Scheduling":[49],"Problem":[50],"(PAISP),":[51],"formal":[53,127],"framework":[54],"where":[55],"loss":[57],"from":[58],"tier-switching":[59],"is":[60,100],"discrete":[61],"and":[62,91,131,150],"model-specific.":[63],"We":[64],"characterize":[65],"safety-quality":[67],"Pareto":[68],"frontier":[69],"as":[70],"fundamental":[72],"structural":[73],"property":[74],"PAISP,":[76],"showing":[77],"guaranteeing":[80],"battery":[81,128],"safety":[82,98,129,159],"under":[83,147],"non-stationary":[84],"form":[86],"appropriate":[88],"evaluation":[89],"domain,":[90],"scarcity":[94],"rather":[95],"than":[96],"constraint":[99],"binding":[102],"bottleneck.":[104],"To":[105],"address":[106],"this,":[107],"we":[108],"propose":[109],"LEAPS":[110,124,156,179],"(Lightweight":[111],"EH-Adaptive":[112],"Policy":[113],"Scheduler),":[114],"employs":[116],"an":[117],"EH-Proportional":[118],"Opportunity":[119],"Cost":[120],"(EPOC)":[121],"scoring":[122],"rule.":[123],"provides":[125],"certificate":[130],"adapts":[132],"selection":[134],"to":[135],"real-time":[136],"conditions":[138,153],"without":[139,184],"manual":[140],"threshold":[141],"calibration.":[142],"Simulations":[143],"over":[144],"21,600":[145],"slots":[146],"diverse":[148],"seasonal":[149],"weather-varying":[151],"confirm":[154],"maintains":[157],"zero":[158],"violations":[160],"while":[161],"matching":[162],"exceeding":[164],"Average":[166],"Quality":[167],"(AQ)":[168],"best":[171],"manually-tuned":[172],"safe":[173],"policy.":[174],"Across":[175],"multiple":[176],"scenarios,":[178],"consistently":[180],"delivers":[181],"competitive":[182],"parameter":[185],"re-tuning,":[186],"demonstrating":[187],"its":[188],"deployment":[189],"flexibility":[190],"for":[191],"resource-constrained":[192],"platforms.":[195]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-08-25T00:00:00"}