{"id":"https://openalex.org/W4415250915","doi":"https://doi.org/10.1109/hpec67600.2025.11196093","title":"Towards -O<sub>mL</sub>: A Deep Learning Based Approach for Outperforming Compiler Defaults","display_name":"Towards -O<sub>mL</sub>: A Deep Learning Based Approach for Outperforming Compiler Defaults","publication_year":2025,"publication_date":"2025-09-15","ids":{"openalex":"https://openalex.org/W4415250915","doi":"https://doi.org/10.1109/hpec67600.2025.11196093"},"language":"en","primary_location":{"id":"doi:10.1109/hpec67600.2025.11196093","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec67600.2025.11196093","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"},"type":"article","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/A5014027237","display_name":"Hafsah Shahzad","orcid":"https://orcid.org/0000-0003-1181-5603"},"institutions":[{"id":"https://openalex.org/I111088046","display_name":"Boston University","ror":"https://ror.org/05qwgg493","country_code":"US","type":"education","lineage":["https://openalex.org/I111088046"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Hafsah Shahzad","raw_affiliation_strings":["Boston University,CAAD Lab,ECE Department"],"affiliations":[{"raw_affiliation_string":"Boston University,CAAD Lab,ECE Department","institution_ids":["https://openalex.org/I111088046"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5034744458","display_name":"Ahmed Sanaullah","orcid":"https://orcid.org/0000-0002-7169-480X"},"institutions":[{"id":"https://openalex.org/I1328219102","display_name":"Red Hat (United States)","ror":"https://ror.org/03ncs3316","country_code":"US","type":"company","lineage":["https://openalex.org/I1328219102"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ahmed Sanaullah","raw_affiliation_strings":["Red Hat Inc"],"affiliations":[{"raw_affiliation_string":"Red Hat Inc","institution_ids":["https://openalex.org/I1328219102"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106133528","display_name":"S. Arora","orcid":null},"institutions":[{"id":"https://openalex.org/I1328219102","display_name":"Red Hat (United States)","ror":"https://ror.org/03ncs3316","country_code":"US","type":"company","lineage":["https://openalex.org/I1328219102"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sanjay Arora","raw_affiliation_strings":["Red Hat Inc"],"affiliations":[{"raw_affiliation_string":"Red Hat Inc","institution_ids":["https://openalex.org/I1328219102"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006985358","display_name":"Ulrich Drepper","orcid":"https://orcid.org/0009-0004-5120-7869"},"institutions":[{"id":"https://openalex.org/I1328219102","display_name":"Red Hat (United States)","ror":"https://ror.org/03ncs3316","country_code":"US","type":"company","lineage":["https://openalex.org/I1328219102"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ulrich Drepper","raw_affiliation_strings":["Red Hat Inc"],"affiliations":[{"raw_affiliation_string":"Red Hat Inc","institution_ids":["https://openalex.org/I1328219102"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021051610","display_name":"Martin Herbordt","orcid":"https://orcid.org/0000-0002-3443-9113"},"institutions":[{"id":"https://openalex.org/I111088046","display_name":"Boston University","ror":"https://ror.org/05qwgg493","country_code":"US","type":"education","lineage":["https://openalex.org/I111088046"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Martin Herbordt","raw_affiliation_strings":["Boston University,CAAD Lab,ECE Department"],"affiliations":[{"raw_affiliation_string":"Boston University,CAAD Lab,ECE Department","institution_ids":["https://openalex.org/I111088046"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5014027237"],"corresponding_institution_ids":["https://openalex.org/I111088046"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.36701007,"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/T12127","display_name":"Software System Performance and Reliability","score":0.9865000247955322,"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/T12127","display_name":"Software System Performance and Reliability","score":0.9865000247955322,"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/T10260","display_name":"Software Engineering Research","score":0.9740999937057495,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11122","display_name":"Online Learning and Analytics","score":0.9643999934196472,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/compiler","display_name":"Compiler","score":0.7950999736785889},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.6729000210762024},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.6026999950408936},{"id":"https://openalex.org/keywords/byte","display_name":"Byte","score":0.46459999680519104},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.4544999897480011},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.4526999890804291},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4474000036716461}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8281000256538391},{"id":"https://openalex.org/C169590947","wikidata":"https://www.wikidata.org/wiki/Q47506","display_name":"Compiler","level":2,"score":0.7950999736785889},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.6729000210762024},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.6026999950408936},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5142999887466431},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.46459999680519104},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.4544999897480011},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.4526999890804291},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4474000036716461},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.37459999322891235},{"id":"https://openalex.org/C133162039","wikidata":"https://www.wikidata.org/wiki/Q1061077","display_name":"Code generation","level":3,"score":0.36320000886917114},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.3215999901294708},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.31380000710487366},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.2892000079154968},{"id":"https://openalex.org/C69637215","wikidata":"https://www.wikidata.org/wiki/Q702362","display_name":"Default","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2777000069618225},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.2522999942302704},{"id":"https://openalex.org/C190902152","wikidata":"https://www.wikidata.org/wiki/Q1325106","display_name":"Optimizing compiler","level":3,"score":0.251800000667572}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/hpec67600.2025.11196093","is_oa":false,"landing_page_url":"https://doi.org/10.1109/hpec67600.2025.11196093","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE High Performance Extreme Computing Conference (HPEC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":35,"referenced_works":["https://openalex.org/W350392096","https://openalex.org/W1991495385","https://openalex.org/W2137742773","https://openalex.org/W2156560068","https://openalex.org/W2157310776","https://openalex.org/W2503699867","https://openalex.org/W2751901133","https://openalex.org/W2780361452","https://openalex.org/W2783614532","https://openalex.org/W3006102157","https://openalex.org/W3008788679","https://openalex.org/W3033154127","https://openalex.org/W3082673232","https://openalex.org/W3082811184","https://openalex.org/W3084442299","https://openalex.org/W3114760331","https://openalex.org/W3118280259","https://openalex.org/W3130095569","https://openalex.org/W3160539805","https://openalex.org/W3163661296","https://openalex.org/W3174413148","https://openalex.org/W3200950867","https://openalex.org/W3207198460","https://openalex.org/W4220727415","https://openalex.org/W4220880378","https://openalex.org/W4318603223","https://openalex.org/W4388483669","https://openalex.org/W4390188155","https://openalex.org/W4390740757","https://openalex.org/W4391951286","https://openalex.org/W4392265924","https://openalex.org/W4396949821","https://openalex.org/W4399851369","https://openalex.org/W4409132205","https://openalex.org/W4413966977"],"related_works":[],"abstract_inverted_index":{"Compilers":[0],"offer":[1],"default":[2,35],"optimization":[3,235],"levels":[4,36],"(e.g.,":[5],"-O<inf":[6,9,96,134,175,198,232],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[7,10,97,114,135,176,199,233],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">z</inf>,":[8],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">3</inf>)":[11],"to":[12,85,145,197,206,214],"generate":[13,38],"high":[14],"performance":[15,54,76,193,217],"code":[16,39,110,127,169],"based":[17,120],"on":[18,172],"developer":[19],"goals":[20],"such":[21,104,216],"as":[22,91,143],"size,":[23],"speed,":[24],"or":[25],"energy":[26],"efficiency.":[27],"As":[28,181],"prior":[29],"work":[30],"has":[31],"shown,":[32],"however,":[33,186],"these":[34],"often":[37],"that":[40],"leaves":[41],"substantial":[42],"room":[43],"for":[44,109,178],"improvement.":[45],"While":[46],"per-application":[47],"tuning":[48],"of":[49,77,129,153,162,174,227],"compiler":[50,78,90],"heuristics":[51],"can":[52,73],"yield":[53],"benefits,":[55],"it":[56],"is":[57,65,82],"time-consuming":[58],"and":[59,80,137,230],"lacks":[60],"generality.":[61],"A":[62],"preferred":[63],"solution":[64],"one":[66],"where":[67],"a":[68,105,149,204,228],"single":[69],"deep":[70,106],"learning":[71,107,118],"model":[72,108,159,213],"(i)":[74],"surpass":[75],"defaults":[79],"(ii)":[81],"sufficiently":[83],"practical":[84,229],"be":[86,207],"integrated":[87],"into":[88],"the":[89,163,212,225],"its":[92],"own":[93],"option,":[94],"e.g.,":[95],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">mL</inf>.In":[98],"this":[99,158],"work,":[100],"we":[101],"first":[102],"train":[103],"size":[111,128,170],"reduction":[112,125,142,171],"(-O<inf":[113],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">zmL</inf>).":[115],"Our":[116],"reinforcement":[117],"(RL)":[119],"approach":[121],"achieves":[122],"an":[123,166],"average":[124,167],"in":[126,209,237],"1.40%":[130],"over":[131],"GCC":[132],"13\u2019s":[133],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">z</inf>":[136,177],"yields":[138],"1.37%":[139],"better":[140],"byte":[141],"compared":[144,196],"state-of-the-art":[146],"efforts.":[147],"Across":[148],"large,":[150],"diverse":[151],"set":[152],"functions":[154,164,191],"from":[155],"standard":[156],"benchmarks,":[157],"optimizes":[160],"36.7%":[161],"achieving":[165],"8.45%":[168],"top":[173],"those":[179],"functions.":[180],"expected":[182],"with":[183,211],"any":[184],"model,":[185],"there":[187],"are":[188],"still":[189],"some":[190],"whose":[192],"degrades":[194],"when":[195],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">z</inf>.":[200],"We":[201],"therefore":[202],"propose":[203],"classifier":[205],"used":[208],"tandem":[210],"reduce":[215],"regressions,":[218],"reducing":[219],"them":[220],"by":[221],"99%.":[222],"Results":[223],"demonstrate":[224],"viability":[226],"learning-based":[231],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">mL</inf>":[234],"level":[236],"production":[238],"compilers.":[239]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-16T00:00:00"}