{"id":"https://openalex.org/W4388650994","doi":"https://doi.org/10.48550/arxiv.2311.05782","title":"MPGemmFI: A Fault Injection Technique for Mixed Precision GEMM in ML Applications","display_name":"MPGemmFI: A Fault Injection Technique for Mixed Precision GEMM in ML Applications","publication_year":2023,"publication_date":"2023-11-09","ids":{"openalex":"https://openalex.org/W4388650994","doi":"https://doi.org/10.48550/arxiv.2311.05782"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2311.05782","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.05782","pdf_url":"https://arxiv.org/pdf/2311.05782","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2311.05782","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101558676","display_name":"Bo Fang","orcid":"https://orcid.org/0009-0003-3847-5765"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Fang, Bo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100370263","display_name":"Xinyi Li","orcid":"https://orcid.org/0000-0001-6379-0335"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Xinyi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058305816","display_name":"Harvey Dam","orcid":"https://orcid.org/0009-0005-9878-7642"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dam, Harvey","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059973546","display_name":"Cheng Tan","orcid":"https://orcid.org/0000-0003-3727-2889"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Cheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5054590774","display_name":"Siva Kumar Sastry Hari","orcid":"https://orcid.org/0000-0001-8346-7981"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hari, Siva Kumar Sastry","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027550136","display_name":"Timothy Tsai","orcid":"https://orcid.org/0000-0001-8190-675X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tsai, Timothy","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033868370","display_name":"Ignacio Laguna","orcid":"https://orcid.org/0000-0002-9374-4433"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Laguna, Ignacio","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063703614","display_name":"Dingwen Tao","orcid":"https://orcid.org/0000-0001-5422-4497"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao, Dingwen","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069846379","display_name":"Ganesh Gopalakrishnan","orcid":"https://orcid.org/0000-0002-3705-0031"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gopalakrishnan, Ganesh","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027932289","display_name":"Prashant Nair","orcid":"https://orcid.org/0000-0002-1349-2273"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nair, Prashant","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038062469","display_name":"Kevin Barker","orcid":"https://orcid.org/0000-0003-4947-0559"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Barker, Kevin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100413657","display_name":"Ang Li","orcid":"https://orcid.org/0000-0003-3734-9137"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Ang","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":12,"corresponding_author_ids":["https://openalex.org/A5101558676"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9995999932289124,"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9990000128746033,"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/T11005","display_name":"Radiation Effects in Electronics","score":0.9932000041007996,"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/computer-science","display_name":"Computer science","score":0.7972966432571411},{"id":"https://openalex.org/keywords/correctness","display_name":"Correctness","score":0.7599458694458008},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.648129940032959},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.6279957890510559},{"id":"https://openalex.org/keywords/computer-engineering","display_name":"Computer engineering","score":0.5205832123756409},{"id":"https://openalex.org/keywords/resilience","display_name":"Resilience (materials science)","score":0.5033401846885681},{"id":"https://openalex.org/keywords/matrix-multiplication","display_name":"Matrix multiplication","score":0.4885343313217163},{"id":"https://openalex.org/keywords/fault-injection","display_name":"Fault injection","score":0.43644753098487854},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.37540769577026367},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.35081788897514343},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.23697012662887573},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.11350440979003906}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7972966432571411},{"id":"https://openalex.org/C55439883","wikidata":"https://www.wikidata.org/wiki/Q360812","display_name":"Correctness","level":2,"score":0.7599458694458008},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.648129940032959},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.6279957890510559},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.5205832123756409},{"id":"https://openalex.org/C2779585090","wikidata":"https://www.wikidata.org/wiki/Q3457762","display_name":"Resilience (materials science)","level":2,"score":0.5033401846885681},{"id":"https://openalex.org/C17349429","wikidata":"https://www.wikidata.org/wiki/Q1049914","display_name":"Matrix multiplication","level":3,"score":0.4885343313217163},{"id":"https://openalex.org/C2775928411","wikidata":"https://www.wikidata.org/wiki/Q2041312","display_name":"Fault injection","level":3,"score":0.43644753098487854},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.37540769577026367},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.35081788897514343},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.23697012662887573},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.11350440979003906},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.0},{"id":"https://openalex.org/C84114770","wikidata":"https://www.wikidata.org/wiki/Q46344","display_name":"Quantum","level":2,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2311.05782","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.05782","pdf_url":"https://arxiv.org/pdf/2311.05782","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2311.05782","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2311.05782","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":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2311.05782","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2311.05782","pdf_url":"https://arxiv.org/pdf/2311.05782","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4388650994.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W1667647204","https://openalex.org/W2404647514","https://openalex.org/W4247536566","https://openalex.org/W2018477250","https://openalex.org/W3119814709","https://openalex.org/W4241418540","https://openalex.org/W1508895727","https://openalex.org/W2725786787","https://openalex.org/W2929170389","https://openalex.org/W2950268673"],"abstract_inverted_index":{"Emerging":[0],"deep":[1],"learning":[2,99],"workloads":[3],"urgently":[4],"need":[5],"fast":[6],"general":[7],"matrix":[8],"multiplication":[9],"(GEMM).":[10],"To":[11,71],"meet":[12],"such":[13,23],"demand,":[14],"one":[15],"of":[16,20,37,97],"the":[17,35,63,85,92,95,103,118,123,135],"critical":[18],"features":[19],"machine-learning-specific":[21],"accelerators":[22],"as":[24],"NVIDIA":[25],"Tensor":[26],"Cores,":[27,30],"AMD":[28],"Matrix":[29],"and":[31,48,57,112],"Google":[32],"TPUs":[33],"is":[34],"support":[36],"mixed-precision":[38,64,86],"enabled":[39],"GEMM.":[40],"For":[41],"DNN":[42],"models,":[43],"lower-precision":[44],"FP":[45],"data":[46],"formats":[47],"computation":[49,65,87],"offer":[50,108],"acceptable":[51],"correctness":[52],"but":[53],"significant":[54],"performance,":[55],"area,":[56],"memory":[58],"footprint":[59],"improvement.":[60],"While":[61],"promising,":[62],"on":[66,102],"error":[67,104,110],"resilience":[68,105],"remains":[69],"unexplored.":[70],"this":[72],"end,":[73],"we":[74,107],"develop":[75],"a":[76],"fault":[77,83],"injection":[78],"framework":[79],"that":[80,115],"systematically":[81],"injects":[82],"into":[84,134],"results.":[88],"We":[89],"investigate":[90],"how":[91],"faults":[93],"affect":[94],"accuracy":[96,121],"machine":[98],"applications.":[100],"Based":[101],"characteristics,":[106],"lightweight":[109],"detection":[111],"correction":[113],"solutions":[114,129],"significantly":[116],"improve":[117],"overall":[119],"model":[120],"if":[122],"models":[124],"experience":[125],"hardware":[126],"faults.":[127],"The":[128],"can":[130],"be":[131],"efficiently":[132],"integrated":[133],"accelerator's":[136],"pipelines.":[137]},"counts_by_year":[],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2023-11-14T00:00:00"}