{"id":"https://openalex.org/W4312121107","doi":"https://doi.org/10.1145/3508352.3549329","title":"Combining Gradients and Probabilities for Heterogeneous Approximation of Neural Networks","display_name":"Combining Gradients and Probabilities for Heterogeneous Approximation of Neural Networks","publication_year":2022,"publication_date":"2022-10-30","ids":{"openalex":"https://openalex.org/W4312121107","doi":"https://doi.org/10.1145/3508352.3549329"},"language":"en","primary_location":{"id":"doi:10.1145/3508352.3549329","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3508352.3549329","pdf_url":null,"source":{"id":"https://openalex.org/S4363608844","display_name":"Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design","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/A5021279961","display_name":"Elias Trommer","orcid":"https://orcid.org/0000-0002-5671-3291"},"institutions":[{"id":"https://openalex.org/I137594350","display_name":"Infineon Technologies (Germany)","ror":"https://ror.org/005kw6t15","country_code":"DE","type":"company","lineage":["https://openalex.org/I137594350"]}],"countries":["DE"],"is_corresponding":true,"raw_author_name":"Elias Trommer","raw_affiliation_strings":["Infineon Technologies, Dresden, Germany"],"affiliations":[{"raw_affiliation_string":"Infineon Technologies, Dresden, Germany","institution_ids":["https://openalex.org/I137594350"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067647232","display_name":"Bernd Waschneck","orcid":"https://orcid.org/0000-0003-0294-8594"},"institutions":[{"id":"https://openalex.org/I137594350","display_name":"Infineon Technologies (Germany)","ror":"https://ror.org/005kw6t15","country_code":"DE","type":"company","lineage":["https://openalex.org/I137594350"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Bernd Waschneck","raw_affiliation_strings":["Infineon Technologies, Dresden, Germany"],"affiliations":[{"raw_affiliation_string":"Infineon Technologies, Dresden, Germany","institution_ids":["https://openalex.org/I137594350"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100755285","display_name":"Akash Kumar","orcid":"https://orcid.org/0000-0001-7125-1737"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Akash Kumar","raw_affiliation_strings":["Center for Advancing Electronics Dresden (cfaed), Dresden, Germany"],"affiliations":[{"raw_affiliation_string":"Center for Advancing Electronics Dresden (cfaed), Dresden, Germany","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5021279961"],"corresponding_institution_ids":["https://openalex.org/I137594350"],"apc_list":null,"apc_paid":null,"fwci":0.1799,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.52074718,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":96},"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":0.9997000098228455,"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":0.9997000098228455,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.9969000220298767,"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"}},{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9929999709129333,"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/artificial-neural-network","display_name":"Artificial neural network","score":0.6593320965766907},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.622176468372345},{"id":"https://openalex.org/keywords/backpropagation","display_name":"Backpropagation","score":0.5926578640937805},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.5102465152740479},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.5017430782318115},{"id":"https://openalex.org/keywords/multiplier","display_name":"Multiplier (economics)","score":0.4968302547931671},{"id":"https://openalex.org/keywords/gaussian-noise","display_name":"Gaussian noise","score":0.4928452968597412},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4664468467235565},{"id":"https://openalex.org/keywords/approximation-error","display_name":"Approximation error","score":0.4612945318222046},{"id":"https://openalex.org/keywords/heuristic","display_name":"Heuristic","score":0.4277102053165436},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.33483877778053284},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.31611213088035583},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.2734466791152954}],"concepts":[{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.6593320965766907},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.622176468372345},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.5926578640937805},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5102465152740479},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.5017430782318115},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.4968302547931671},{"id":"https://openalex.org/C4199805","wikidata":"https://www.wikidata.org/wiki/Q2725903","display_name":"Gaussian noise","level":2,"score":0.4928452968597412},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4664468467235565},{"id":"https://openalex.org/C122383733","wikidata":"https://www.wikidata.org/wiki/Q865920","display_name":"Approximation error","level":2,"score":0.4612945318222046},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.4277102053165436},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.33483877778053284},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.31611213088035583},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2734466791152954},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"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/C139719470","wikidata":"https://www.wikidata.org/wiki/Q39680","display_name":"Macroeconomics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3508352.3549329","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3508352.3549329","pdf_url":null,"source":{"id":"https://openalex.org/S4363608844","display_name":"Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design","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":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.9100000262260437}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W2051831389","https://openalex.org/W2108598243","https://openalex.org/W2533121491","https://openalex.org/W2560674852","https://openalex.org/W2612139336","https://openalex.org/W2619638368","https://openalex.org/W2796625795","https://openalex.org/W2798993323","https://openalex.org/W2804268686","https://openalex.org/W2958306322","https://openalex.org/W2963163009","https://openalex.org/W2990982501","https://openalex.org/W3007372856","https://openalex.org/W3036274717","https://openalex.org/W3036417437","https://openalex.org/W3042701997","https://openalex.org/W3083485280","https://openalex.org/W3091041533","https://openalex.org/W3101791909","https://openalex.org/W3103768881","https://openalex.org/W3128097214","https://openalex.org/W3214870808","https://openalex.org/W4236363946"],"related_works":["https://openalex.org/W2088845016","https://openalex.org/W589102260","https://openalex.org/W1966421350","https://openalex.org/W1868434454","https://openalex.org/W4366985237","https://openalex.org/W2810569973","https://openalex.org/W2128396103","https://openalex.org/W4366984740","https://openalex.org/W4367299891","https://openalex.org/W1518564963"],"abstract_inverted_index":{"This":[0],"work":[1],"explores":[2],"the":[3,23,51,56,74,81,87,101,108,111,113,116,120,127,140,150,164,177,213,220,223],"search":[4],"for":[5,10,39,76,153,159],"heterogeneous":[6,143],"approximate":[7,43,121,217],"multiplier":[8,102,122,218],"configurations":[9],"neural":[11,32,146],"networks":[12],"that":[13,66,139,207],"produce":[14],"high":[15,232],"accuracy":[16,170,199],"and":[17,47,145],"low":[18],"energy":[19,151,192],"consumption.":[20],"We":[21,204],"discuss":[22],"validity":[24],"of":[25,42,50,70,83,100,119,142,200,215,222],"additive":[26,57,128,226],"Gaussian":[27,58,129,227],"noise":[28,59,84,130,228],"added":[29],"to":[30,106,132,157],"accurate":[31,88],"network":[33,93,147],"computations":[34,89],"as":[35,63],"a":[36,64,168,187,195],"surrogate":[37],"model":[38,60,99,114,185,210,230],"behavioral":[40],"simulation":[41],"multipliers.":[44],"The":[45],"continuous":[46],"differentiable":[48],"properties":[49],"solution":[52],"space":[53,131],"spanned":[54],"by":[55,155],"are":[61],"used":[62,225],"heuristic":[65],"generates":[67],"meaningful":[68],"estimates":[69,115],"layer":[71],"robustness":[72],"without":[73],"need":[75],"combinatorial":[77],"optimization":[78],"techniques.":[79],"Instead,":[80],"amount":[82],"injected":[85],"into":[86],"is":[90,104,237],"learned":[91],"during":[92],"training":[94],"using":[95],"backpropagation.":[96],"A":[97],"probabilistic":[98],"error":[103,209],"presented":[105],"bridge":[107],"gap":[109],"between":[110],"domains;":[112],"standard":[117],"deviation":[118],"error,":[123],"connecting":[124],"solutions":[125],"in":[126,191,197,219],"actual":[133],"hardware":[134],"instances.":[135],"Our":[136,234],"experiments":[137],"show":[138],"combination":[141],"approximation":[144],"retraining":[148],"reduces":[149],"consumption":[152,193],"multiplications":[154],"70%":[156],"79%":[158],"different":[160],"ResNet":[161],"variants":[162],"on":[163],"CIFAR-10":[165],"dataset":[166],"with":[167,194,231],"Top-1":[169],"loss":[171],"below":[172],"one":[173],"percentage":[174,202],"point.":[175],"For":[176],"more":[178],"complex":[179],"Tiny":[180],"ImageNet":[181],"task,":[182],"our":[183,208],"VGG16":[184],"achieves":[186],"53":[188],"%":[189],"reduction":[190],"drop":[196],"Top-5":[198],"0.5":[201],"points.":[203],"further":[205],"demonstrate":[206],"can":[211],"predict":[212],"parameters":[214],"an":[216],"context":[221],"commonly":[224],"(AGN)":[229],"accuracy.":[233],"software":[235],"implementation":[236],"available":[238],"under":[239],"https://github.com/etrommer/agn-approx.":[240]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2023,"cited_by_count":2}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}