{"id":"https://openalex.org/W7165120350","doi":"https://doi.org/10.1145/3787109.3816404","title":"Shallow Enough? A Cross-Architecture Study of Ultra-Low-Depth Neural Networks for Edge Inference","display_name":"Shallow Enough? A Cross-Architecture Study of Ultra-Low-Depth Neural Networks for Edge Inference","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165120350","doi":"https://doi.org/10.1145/3787109.3816404"},"language":null,"primary_location":{"id":"doi:10.1145/3787109.3816404","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3816404","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3787109.3816404","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5118662884","display_name":"Chengwei Zhou","orcid":null},"institutions":[{"id":"https://openalex.org/I58956616","display_name":"Case Western Reserve University","ror":"https://ror.org/051fd9666","country_code":"US","type":"education","lineage":["https://openalex.org/I58956616"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chengwei Zhou","raw_affiliation_strings":["Case Western Reserve University, CLEVELAND, OH, USA"],"raw_orcid":"https://orcid.org/0009-0000-4539-5681","affiliations":[{"raw_affiliation_string":"Case Western Reserve University, CLEVELAND, OH, USA","institution_ids":["https://openalex.org/I58956616"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137912823","display_name":"H Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I58956616","display_name":"Case Western Reserve University","ror":"https://ror.org/051fd9666","country_code":"US","type":"education","lineage":["https://openalex.org/I58956616"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Haotian Yu","raw_affiliation_strings":["Case Western Reserve University, CLEVELAND, Ohio, USA"],"raw_orcid":"https://orcid.org/0009-0004-4057-1286","affiliations":[{"raw_affiliation_string":"Case Western Reserve University, CLEVELAND, Ohio, USA","institution_ids":["https://openalex.org/I58956616"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138860076","display_name":"Shoma Yukawa","orcid":"https://orcid.org/0009-0006-8488-3008"},"institutions":[{"id":"https://openalex.org/I58956616","display_name":"Case Western Reserve University","ror":"https://ror.org/051fd9666","country_code":"US","type":"education","lineage":["https://openalex.org/I58956616"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shoma Yukawa","raw_affiliation_strings":["Case Western Reserve University, CLEVELAND, Ohio, USA"],"raw_orcid":"https://orcid.org/0009-0006-8488-3008","affiliations":[{"raw_affiliation_string":"Case Western Reserve University, CLEVELAND, Ohio, USA","institution_ids":["https://openalex.org/I58956616"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078184577","display_name":"Deniz Najafi","orcid":"https://orcid.org/0009-0008-2734-8935"},"institutions":[{"id":"https://openalex.org/I118118575","display_name":"New Jersey Institute of Technology","ror":"https://ror.org/05e74xb87","country_code":"US","type":"education","lineage":["https://openalex.org/I118118575"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Deniz Najafi","raw_affiliation_strings":["ECE, New Jersey Institute of Technology, Newark, NJ, USA"],"raw_orcid":"https://orcid.org/0009-0008-2734-8935","affiliations":[{"raw_affiliation_string":"ECE, New Jersey Institute of Technology, Newark, NJ, USA","institution_ids":["https://openalex.org/I118118575"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051680668","display_name":"Shaahin Angizi","orcid":"https://orcid.org/0000-0003-2289-6381"},"institutions":[{"id":"https://openalex.org/I118118575","display_name":"New Jersey Institute of Technology","ror":"https://ror.org/05e74xb87","country_code":"US","type":"education","lineage":["https://openalex.org/I118118575"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shaahin Angizi","raw_affiliation_strings":["New Jersey Institute of Technology, Newark, NJ, USA"],"raw_orcid":"https://orcid.org/0000-0003-2289-6381","affiliations":[{"raw_affiliation_string":"New Jersey Institute of Technology, Newark, NJ, USA","institution_ids":["https://openalex.org/I118118575"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5017435097","display_name":"Gourav Datta","orcid":"https://orcid.org/0000-0002-5380-2619"},"institutions":[{"id":"https://openalex.org/I58956616","display_name":"Case Western Reserve University","ror":"https://ror.org/051fd9666","country_code":"US","type":"education","lineage":["https://openalex.org/I58956616"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Gourav Datta","raw_affiliation_strings":["Case Western Reserve University, CLEVELAND, OH, USA"],"raw_orcid":"https://orcid.org/0000-0002-5380-2619","affiliations":[{"raw_affiliation_string":"Case Western Reserve University, CLEVELAND, OH, USA","institution_ids":["https://openalex.org/I58956616"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"220","last_page":"225"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.8320000171661377,"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":0.8320000171661377,"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.029999999329447746,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.01640000008046627,"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/software-deployment","display_name":"Software deployment","score":0.5870000123977661},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5360999703407288},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5216000080108643},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.5149999856948853},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4921000003814697},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.4830000102519989},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.4487999975681305},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.4440999925136566},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.4377000033855438}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7042999863624573},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.5870000123977661},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5360999703407288},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5216000080108643},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.5149999856948853},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4921000003814697},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.4830000102519989},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.4487999975681305},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.4440999925136566},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.4377000033855438},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42980000376701355},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.40369999408721924},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3887999951839447},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3801000118255615},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.35530000925064087},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.3513000011444092},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3467000126838684},{"id":"https://openalex.org/C193415008","wikidata":"https://www.wikidata.org/wiki/Q639681","display_name":"Network architecture","level":2,"score":0.3458999991416931},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32190001010894775},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3050000071525574},{"id":"https://openalex.org/C177225278","wikidata":"https://www.wikidata.org/wiki/Q192674","display_name":"Factoring","level":2,"score":0.2978000044822693},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.29589998722076416},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.29159998893737793},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.2892000079154968},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.28209999203681946},{"id":"https://openalex.org/C134261354","wikidata":"https://www.wikidata.org/wiki/Q938438","display_name":"Statistical inference","level":2,"score":0.28130000829696655},{"id":"https://openalex.org/C76752949","wikidata":"https://www.wikidata.org/wiki/Q7607499","display_name":"Stencil","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2533999979496002}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3787109.3816404","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3816404","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3787109.3816404","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3787109.3816404","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 Great Lakes Symposium on VLSI 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.4796278178691864,"display_name":"Affordable and clean energy","id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W2752849906","https://openalex.org/W2963122961","https://openalex.org/W2963163009","https://openalex.org/W2964137095","https://openalex.org/W4312671791","https://openalex.org/W4313156423","https://openalex.org/W4390872447","https://openalex.org/W4390874575"],"related_works":[],"abstract_inverted_index":{"Edge":[0],"deployment":[1,111],"imposes":[2],"strict":[3],"latency,":[4,122],"memory,":[5],"and":[6,46,77,91,95,101,109,123,149],"energy":[7,102],"constraints":[8],"that":[9],"scale":[10,125],"directly":[11],"with":[12,80,136],"network":[13],"depth,":[14,76],"yet":[15],"the":[16,23,38,61],"question":[17],"of":[18,42,64,75],"which":[19,140],"architectural":[20],"paradigm":[21],"offers":[22],"best":[24],"accuracy-efficiency":[25],"tradeoff":[26],"at":[27],"ultra-low":[28],"depth":[29,53,132,148],"(four":[30],"to":[31,143],"six":[32],"layers)":[33],"remains":[34],"open.":[35],"We":[36,55],"present":[37],"first":[39],"systematic":[40],"comparison":[41],"convolutional,":[43],"pure":[44],"transformer,":[45],"hybrid":[47],"CNN-transformer":[48],"models":[49],"under":[50],"fixed":[51],"shallow":[52],"budgets.":[54],"develop":[56],"an":[57,105],"analytical":[58],"framework":[59],"characterizing":[60],"representational":[62],"cost":[63],"local":[65],"convolutional":[66],"versus":[67],"global":[68],"attention-based":[69],"feature":[70],"mixing":[71],"as":[72,131],"a":[73,146],"function":[74],"validate":[78],"it":[79],"experiments":[81,118],"on":[82,104,113,139],"ImageNet-1K":[83],"across":[84,126],"architectures":[85],"spanning":[86],"MobileNetV2,":[87],"DeiT,":[88],"Swin,":[89],"MobileViT,":[90],"EfficientFormer.":[92],"Beyond":[93],"FLOPs":[94],"accuracy,":[96,121],"we":[97],"report":[98],"real-world":[99],"latency":[100],"measurements":[103],"NVIDIA":[106],"Jetson":[107],"Nano":[108],"examine":[110],"feasibility":[112],"Cortex-M":[114],"class":[115,142],"microcontrollers.":[116],"Our":[117],"reveal":[119],"how":[120],"memory":[124],"all":[127],"three":[128],"architecture":[129],"families":[130],"decreases,":[133],"providing":[134],"practitioners":[135],"direct":[137],"guidance":[138],"model":[141],"choose":[144],"for":[145],"given":[147],"hardware":[150],"budget.":[151]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-06-19T00:00:00"}