{"id":"https://openalex.org/W4416203749","doi":"https://doi.org/10.1145/3712285.3759900","title":"PerfDojo: Automated ML Library Generation for Heterogeneous Architectures","display_name":"PerfDojo: Automated ML Library Generation for Heterogeneous Architectures","publication_year":2025,"publication_date":"2025-11-12","ids":{"openalex":"https://openalex.org/W4416203749","doi":"https://doi.org/10.1145/3712285.3759900"},"language":"en","primary_location":{"id":"doi:10.1145/3712285.3759900","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759900","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3712285.3759900","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101942908","display_name":"Andrei Ivanov","orcid":"https://orcid.org/0009-0007-9487-9990"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Andrei Ivanov","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107860493","display_name":"Siyuan Shen","orcid":"https://orcid.org/0009-0003-0061-5072"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Siyuan Shen","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120351180","display_name":"Gioele Gottardo","orcid":"https://orcid.org/0009-0009-9103-1403"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Gioele Gottardo","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001468122","display_name":"Marcin Chrapek","orcid":"https://orcid.org/0009-0009-7654-6038"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Marcin Chrapek","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120447623","display_name":"Afif Boudaoud","orcid":"https://orcid.org/0009-0003-8662-6353"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Afif Boudaoud","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066066810","display_name":"Timo Schneider","orcid":"https://orcid.org/0000-0002-4884-3934"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Timo Schneider","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043408422","display_name":"Luca Benini","orcid":"https://orcid.org/0000-0001-8068-3806"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Luca Benini","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026990786","display_name":"Torsten Hoefler","orcid":"https://orcid.org/0000-0002-1333-9797"},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Torsten Hoefler","raw_affiliation_strings":["ETH Z\u00fcrich, Zurich, Switzerland"],"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich, Zurich, Switzerland","institution_ids":["https://openalex.org/I35440088"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5101942908"],"corresponding_institution_ids":["https://openalex.org/I35440088"],"apc_list":null,"apc_paid":null,"fwci":2.5269,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.91442491,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"137","last_page":"151"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.4334999918937683,"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":0.4334999918937683,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.0989999994635582,"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":0.08910000324249268,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.6172999739646912},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5432999730110168},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.4307999908924103},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.40720000863075256},{"id":"https://openalex.org/keywords/source-code","display_name":"Source code","score":0.37709999084472656},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.37529999017715454}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8398000001907349},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.6172999739646912},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5432999730110168},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45089998841285706},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.4307999908924103},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.40720000863075256},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3837999999523163},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.37709999084472656},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.37529999017715454},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C133162039","wikidata":"https://www.wikidata.org/wiki/Q1061077","display_name":"Code generation","level":3,"score":0.31310001015663147},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.2957000136375427},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.28189998865127563},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.27549999952316284},{"id":"https://openalex.org/C139571649","wikidata":"https://www.wikidata.org/wiki/Q1156793","display_name":"Program optimization","level":3,"score":0.27480000257492065},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.2651999890804291},{"id":"https://openalex.org/C169087156","wikidata":"https://www.wikidata.org/wiki/Q2131593","display_name":"Framing (construction)","level":2,"score":0.26429998874664307},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.2554999887943268}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1145/3712285.3759900","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759900","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2511.03586","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2511.03586","pdf_url":"https://arxiv.org/pdf/2511.03586","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:doi:10.3929/ethz-c-000789873","is_oa":true,"landing_page_url":"http://hdl.handle.net/20.500.11850/789873","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference Paper"},{"id":"pmh:oai:cris.unibo.it:11585/1040873","is_oa":false,"landing_page_url":"https://hdl.handle.net/11585/1040873","pdf_url":null,"source":{"id":"https://openalex.org/S4306402579","display_name":"Archivio istituzionale della ricerca (Alma Mater Studiorum Universit\u00e0 di Bologna)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210117483","host_organization_name":"Istituto di Ematologia di Bologna","host_organization_lineage":["https://openalex.org/I4210117483"],"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":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.1145/3712285.3759900","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3712285.3759900","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W1557517019","https://openalex.org/W2034761517","https://openalex.org/W2046376809","https://openalex.org/W2145339207","https://openalex.org/W2170901118","https://openalex.org/W2918037051","https://openalex.org/W2949967139","https://openalex.org/W2954698171","https://openalex.org/W2962993073","https://openalex.org/W2979365412","https://openalex.org/W2987684178","https://openalex.org/W2994193159","https://openalex.org/W3012249773","https://openalex.org/W3013692244","https://openalex.org/W3016309475","https://openalex.org/W3122286897","https://openalex.org/W3130554079","https://openalex.org/W4220727415","https://openalex.org/W4240117034","https://openalex.org/W4251637954","https://openalex.org/W4255653201","https://openalex.org/W4281707342","https://openalex.org/W4300892457","https://openalex.org/W4317810581","https://openalex.org/W4318541538","https://openalex.org/W4380874786","https://openalex.org/W4386738216","https://openalex.org/W4388094337","https://openalex.org/W4393160302","https://openalex.org/W4407217868","https://openalex.org/W4408029577","https://openalex.org/W4415797154"],"related_works":[],"abstract_inverted_index":{"The":[0],"increasing":[1],"complexity":[2],"of":[3,10],"machine":[4],"learning":[5],"models":[6],"and":[7,35,41,61,81,122,141],"the":[8],"proliferation":[9],"diverse":[11,136],"hardware":[12,116],"architectures":[13],"(CPUs,":[14],"GPUs,":[15],"accelerators)":[16],"make":[17],"achieving":[18],"optimal":[19],"performance":[20,45,66,133],"a":[21,71,99],"significant":[22,132],"challenge.":[23],"Heterogeneity":[24],"in":[25],"instruction":[26],"sets,":[27],"specialized":[28],"kernel":[29],"requirements":[30],"for":[31],"different":[32],"data":[33],"types":[34],"model":[36],"features":[37],"(e.g.,":[38],"sparsity,":[39],"quantization),":[40],"architecture-specific":[42],"optimizations":[43],"complicate":[44],"tuning.":[46],"Manual":[47],"optimization":[48,74,93,113],"is":[49,88],"resource-intensive,":[50],"while":[51],"existing":[52],"automatic":[53,73],"approaches":[54],"often":[55],"rely":[56],"on":[57],"complex":[58],"hardware-specific":[59],"heuristics":[60],"uninterpretable":[62],"intermediate":[63],"representations,":[64],"hindering":[65],"portability.":[67],"We":[68,126],"introduce":[69],"PerfLLM,":[70],"novel":[72],"methodology":[75],"leveraging":[76],"Large":[77],"Language":[78],"Models":[79],"(LLMs)":[80],"Reinforcement":[82],"Learning":[83],"(RL).":[84],"Central":[85],"to":[86,130],"this":[87],"PerfDojo,":[89],"an":[90,95],"environment":[91],"framing":[92],"as":[94],"RL":[96,123],"game":[97],"using":[98],"human-readable,":[100],"mathematically-inspired":[101],"code":[102],"representation":[103],"that":[104],"guarantees":[105],"semantic":[106],"validity":[107],"through":[108],"transformations.":[109],"This":[110],"allows":[111],"effective":[112],"without":[114],"prior":[115],"knowledge,":[117],"facilitating":[118],"both":[119],"human":[120],"analysis":[121],"agent":[124],"training.":[125],"demonstrate":[127],"PerfLLM\u2019s":[128],"ability":[129],"achieve":[131],"gains":[134],"across":[135],"CPU":[137],"(x86,":[138],"Arm,":[139],"RISC-V)":[140],"GPU":[142],"architectures.":[143]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-11T14:59:36.786465","created_date":"2025-11-07T00:00:00"}