{"id":"https://openalex.org/W4416252152","doi":"https://doi.org/10.1109/ijcnn64981.2025.11229004","title":"EPSILON: Adaptive Fault Mitigation in Approximate Deep Neural Network using Statistical Signatures","display_name":"EPSILON: Adaptive Fault Mitigation in Approximate Deep Neural Network using Statistical Signatures","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4416252152","doi":"https://doi.org/10.1109/ijcnn64981.2025.11229004"},"language":null,"primary_location":{"id":"doi:10.1109/ijcnn64981.2025.11229004","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11229004","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2504.20074","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5070759206","display_name":"Khurram Khalil","orcid":"https://orcid.org/0000-0003-1718-7150"},"institutions":[{"id":"https://openalex.org/I76835614","display_name":"University of Missouri","ror":"https://ror.org/02ymw8z06","country_code":"US","type":"education","lineage":["https://openalex.org/I76835614"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Khurram Khalil","raw_affiliation_strings":["University of Missouri-Columbia,Department of Electrical Engineering and Computer Science,USA"],"affiliations":[{"raw_affiliation_string":"University of Missouri-Columbia,Department of Electrical Engineering and Computer Science,USA","institution_ids":["https://openalex.org/I76835614"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5069684684","display_name":"Khaza Anuarul Hoque","orcid":"https://orcid.org/0000-0002-1625-6479"},"institutions":[{"id":"https://openalex.org/I76835614","display_name":"University of Missouri","ror":"https://ror.org/02ymw8z06","country_code":"US","type":"education","lineage":["https://openalex.org/I76835614"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Khaza Anuarul Hoque","raw_affiliation_strings":["University of Missouri-Columbia,Department of Electrical Engineering and Computer Science,USA"],"affiliations":[{"raw_affiliation_string":"University of Missouri-Columbia,Department of Electrical Engineering and Computer Science,USA","institution_ids":["https://openalex.org/I76835614"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5070759206"],"corresponding_institution_ids":["https://openalex.org/I76835614"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.34042984,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"8"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.3783000111579895,"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"}},"topics":[{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.3783000111579895,"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/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.23440000414848328,"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/T12676","display_name":"Machine Learning and ELM","score":0.06260000169277191,"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/overhead","display_name":"Overhead (engineering)","score":0.5913000106811523},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.5579000115394592},{"id":"https://openalex.org/keywords/statistical-inference","display_name":"Statistical inference","score":0.5181000232696533},{"id":"https://openalex.org/keywords/fault-detection-and-isolation","display_name":"Fault detection and isolation","score":0.5145000219345093},{"id":"https://openalex.org/keywords/fault","display_name":"Fault (geology)","score":0.49799999594688416},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.49380001425743103},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.47440001368522644},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.4083999991416931}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7081999778747559},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5913000106811523},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.5579000115394592},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.5181000232696533},{"id":"https://openalex.org/C152745839","wikidata":"https://www.wikidata.org/wiki/Q5438153","display_name":"Fault detection and isolation","level":3,"score":0.5145000219345093},{"id":"https://openalex.org/C175551986","wikidata":"https://www.wikidata.org/wiki/Q47089","display_name":"Fault (geology)","level":2,"score":0.49799999594688416},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.49380001425743103},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.47440001368522644},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4124999940395355},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.4083999991416931},{"id":"https://openalex.org/C125611927","wikidata":"https://www.wikidata.org/wiki/Q17008131","display_name":"Criticality","level":2,"score":0.3799999952316284},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.36410000920295715},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.3564999997615814},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.3564999997615814},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.33709999918937683},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.33489999175071716},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3230000138282776},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.32280001044273376},{"id":"https://openalex.org/C63540848","wikidata":"https://www.wikidata.org/wiki/Q3140932","display_name":"Fault tolerance","level":2,"score":0.3111000061035156},{"id":"https://openalex.org/C2779696439","wikidata":"https://www.wikidata.org/wiki/Q7512811","display_name":"Signature (topology)","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2754000127315521},{"id":"https://openalex.org/C126953365","wikidata":"https://www.wikidata.org/wiki/Q5438152","display_name":"Fault coverage","level":3,"score":0.27300000190734863},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.2694000005722046},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.26339998841285706},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2554999887943268}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/ijcnn64981.2025.11229004","is_oa":false,"landing_page_url":"https://doi.org/10.1109/ijcnn64981.2025.11229004","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Joint Conference on Neural Networks (IJCNN)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2504.20074","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.20074","pdf_url":"https://arxiv.org/pdf/2504.20074","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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:doi:10.48550/arxiv.2504.20074","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2504.20074","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.20074","pdf_url":"https://arxiv.org/pdf/2504.20074","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W1998305460","https://openalex.org/W2612139336","https://openalex.org/W2808976322","https://openalex.org/W2898009658","https://openalex.org/W2899084402","https://openalex.org/W2958306322","https://openalex.org/W2978384356","https://openalex.org/W3016208860","https://openalex.org/W3080303273","https://openalex.org/W3130607817","https://openalex.org/W3131039502","https://openalex.org/W3161231436","https://openalex.org/W3162727751","https://openalex.org/W3196044848","https://openalex.org/W4252174618","https://openalex.org/W4293232522","https://openalex.org/W4309326585","https://openalex.org/W4320002709","https://openalex.org/W4376461217","https://openalex.org/W4383503739","https://openalex.org/W4385976015","https://openalex.org/W4390145134","https://openalex.org/W4396909909","https://openalex.org/W4400362675","https://openalex.org/W4401942800","https://openalex.org/W4403196715","https://openalex.org/W4404134107"],"related_works":[],"abstract_inverted_index":{"The":[0],"increasing":[1],"adoption":[2],"of":[3,122,133],"approximate":[4,134,140],"computing":[5],"in":[6,20,79,163,169,182],"deep":[7],"neural":[8],"network":[9,103],"accelerators":[10],"(AxDNNs)":[11],"promises":[12],"significant":[13],"energy":[14,131,170],"efficiency":[15],"gains.":[16],"However,":[17],"permanent":[18],"faults":[19],"AxDNNs":[21,181],"can":[22],"severely":[23],"degrade":[24],"their":[25,29],"performance":[26],"compared":[27],"to":[28,101],"accurate":[30],"counterparts":[31],"(AccDNNs).":[32],"Traditional":[33],"fault":[34,75,92,106,151],"detection":[35,76,93],"and":[36,46,69,77,105,125,150,166],"mitigation":[37,78,115],"approaches,":[38],"while":[39,98,128,159],"effective":[40],"for":[41,51,73,178],"AccDNNs,":[42],"introduce":[43,59],"substantial":[44],"overhead":[45],"latency,":[47],"making":[48],"them":[49],"impractical":[50],"energy-constrained":[52],"real-time":[53],"deployment.":[54],"To":[55],"address":[56],"this,":[57],"we":[58],"EPSILON,":[60],"a":[61,84,119,175],"lightweight":[62],"framework":[63,82],"that":[64,89,154],"leverages":[65],"pre-computed":[66],"statistical":[67,120],"signatures":[68],"layer-wise":[70],"importance":[71],"metrics":[72],"efficient":[74],"AxDNNs.":[80],"Our":[81],"introduces":[83],"novel":[85],"non-parametric":[86],"pattern-matching":[87],"algorithm":[88],"enables":[90],"constant-time":[91],"without":[94],"interrupting":[95],"normal":[96],"execution":[97],"dynamically":[99],"adapting":[100],"different":[102],"architectures":[104],"patterns.":[107],"EPSILON":[108,155,173],"maintains":[109,156],"model":[110],"accuracy":[111,158],"by":[112],"intelligently":[113],"adjusting":[114],"strategies":[116],"based":[117],"on":[118],"analysis":[121],"weight":[123],"distribution":[124],"layer":[126],"criticality":[127],"preserving":[129],"the":[130],"benefits":[132],"computing.":[135],"Extensive":[136],"evaluations":[137],"across":[138],"various":[139],"multipliers,":[141],"AxDNN":[142],"architectures,":[143],"popular":[144],"datasets":[145],"(MNIST,":[146],"CIFAR-10,":[147],"CIFAR-100,":[148],"ImageNet-1k),":[149],"scenarios":[152],"demonstrate":[153],"80.05%":[157],"offering":[160],"22%":[161],"improvement":[162,168],"inference":[164],"time":[165],"28%":[167],"efficiency,":[171],"establishing":[172],"as":[174],"practical":[176],"solution":[177],"deploying":[179],"reliable":[180],"safety-critical":[183],"edge":[184],"applications.":[185]},"counts_by_year":[],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2025-11-14T00:00:00"}