{"id":"https://openalex.org/W4308235972","doi":"https://doi.org/10.1109/icip46576.2022.9897462","title":"Stacking More Linear Operations with Orthogonal Regularization to Learn Better","display_name":"Stacking More Linear Operations with Orthogonal Regularization to Learn Better","publication_year":2022,"publication_date":"2022-10-16","ids":{"openalex":"https://openalex.org/W4308235972","doi":"https://doi.org/10.1109/icip46576.2022.9897462"},"language":"en","primary_location":{"id":"doi:10.1109/icip46576.2022.9897462","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip46576.2022.9897462","pdf_url":null,"source":{"id":"https://openalex.org/S4363607719","display_name":"2022 IEEE International Conference on Image Processing (ICIP)","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":"2022 IEEE International Conference on Image Processing (ICIP)","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/A5062221157","display_name":"Weixiang Xu","orcid":"https://orcid.org/0000-0002-3083-794X"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weixiang Xu","raw_affiliation_strings":["Institute of Automation, Chinese Academy of Sciences,NLPR","NLPR, Institute of Automation, Chinese Academy of Sciences","School of Artificial Intelligence, University of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Automation, Chinese Academy of Sciences,NLPR","institution_ids":["https://openalex.org/I19820366"]},{"raw_affiliation_string":"NLPR, Institute of Automation, Chinese Academy of Sciences","institution_ids":["https://openalex.org/I19820366"]},{"raw_affiliation_string":"School of Artificial Intelligence, University of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5001982715","display_name":"Jian Cheng","orcid":"https://orcid.org/0000-0003-1435-673X"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jian Cheng","raw_affiliation_strings":["Institute of Automation, Chinese Academy of Sciences,NLPR","NLPR, Institute of Automation, Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Institute of Automation, Chinese Academy of Sciences,NLPR","institution_ids":["https://openalex.org/I19820366"]},{"raw_affiliation_string":"NLPR, Institute of Automation, Chinese Academy of Sciences","institution_ids":["https://openalex.org/I19820366"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.10683609,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2731","last_page":"2735"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9987999796867371,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9954000115394592,"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/flops","display_name":"FLOPS","score":0.8546302914619446},{"id":"https://openalex.org/keywords/stacking","display_name":"Stacking","score":0.7858328819274902},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.6871861219406128},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6517274975776672},{"id":"https://openalex.org/keywords/convolution","display_name":"Convolution (computer science)","score":0.6450852155685425},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.5964756011962891},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5599638819694519},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.5544278025627136},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.4855627715587616},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.4789638817310333},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.4187943935394287},{"id":"https://openalex.org/keywords/mathematical-optimization","display_name":"Mathematical optimization","score":0.37366753816604614},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3528563380241394},{"id":"https://openalex.org/keywords/applied-mathematics","display_name":"Applied mathematics","score":0.342197060585022},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.23551321029663086},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.18778955936431885},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.14297115802764893},{"id":"https://openalex.org/keywords/mathematical-analysis","display_name":"Mathematical analysis","score":0.06558966636657715}],"concepts":[{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.8546302914619446},{"id":"https://openalex.org/C33347731","wikidata":"https://www.wikidata.org/wiki/Q285210","display_name":"Stacking","level":2,"score":0.7858328819274902},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.6871861219406128},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6517274975776672},{"id":"https://openalex.org/C45347329","wikidata":"https://www.wikidata.org/wiki/Q5166604","display_name":"Convolution (computer science)","level":3,"score":0.6450852155685425},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5964756011962891},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5599638819694519},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.5544278025627136},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4855627715587616},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4789638817310333},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4187943935394287},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.37366753816604614},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3528563380241394},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.342197060585022},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23551321029663086},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.18778955936431885},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.14297115802764893},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.06558966636657715},{"id":"https://openalex.org/C46141821","wikidata":"https://www.wikidata.org/wiki/Q209402","display_name":"Nuclear magnetic resonance","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":1,"locations":[{"id":"doi:10.1109/icip46576.2022.9897462","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip46576.2022.9897462","pdf_url":null,"source":{"id":"https://openalex.org/S4363607719","display_name":"2022 IEEE International Conference on Image Processing (ICIP)","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":"2022 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W2125930537","https://openalex.org/W2134192361","https://openalex.org/W2279098554","https://openalex.org/W2587753047","https://openalex.org/W2593110912","https://openalex.org/W2752782242","https://openalex.org/W2884585870","https://openalex.org/W2949736877","https://openalex.org/W2963376662","https://openalex.org/W2964047251","https://openalex.org/W2995607862","https://openalex.org/W3034403122","https://openalex.org/W4297779039","https://openalex.org/W4307823382","https://openalex.org/W6638444622","https://openalex.org/W6638523607","https://openalex.org/W6678818196","https://openalex.org/W6695314431","https://openalex.org/W6733068669","https://openalex.org/W6748581759","https://openalex.org/W6755310813","https://openalex.org/W6769906912","https://openalex.org/W7064683377"],"related_works":["https://openalex.org/W2035329725","https://openalex.org/W4315697128","https://openalex.org/W4376641153","https://openalex.org/W2070875936","https://openalex.org/W4250391473","https://openalex.org/W3045075405","https://openalex.org/W4302292679","https://openalex.org/W2995343971","https://openalex.org/W2982536526","https://openalex.org/W2992221004"],"abstract_inverted_index":{"How":[0],"to":[1,25,45,57,88],"improve":[2],"the":[3,14,52,68,90,111,119],"generalization":[4],"of":[5],"CNN":[6,27],"models":[7,28],"has":[8],"been":[9],"a":[10,21,46,59],"long-lasting":[11],"problem":[12],"in":[13],"deep":[15],"learning":[16],"community.":[17],"This":[18],"paper":[19],"presents":[20],"runtime":[22],"parameter/FLOPs-free":[23],"method":[24],"strengthen":[26],"by":[29],"stacking":[30],"linear":[31],"convolution":[32],"operations":[33],"during":[34],"training.":[35],"We":[36],"show":[37],"that":[38,50],"overparameterization":[39],"with":[40],"appropriate":[41],"regularization":[42],"can":[43,107],"lead":[44],"smooth":[47],"optimization":[48],"landscape":[49],"improves":[51],"performance.":[53],"Concretely,":[54],"we":[55],"propose":[56],"add":[58],"1":[60,62,92,94,103,105],"\u00d7":[61,71,93,104,114],"convolutional":[63,73],"layer":[64,74,116],"before":[65],"and":[66],"after":[67],"original":[69,112,120],"k":[70,72,113,115],"respectively,":[75],"without":[76,117],"any":[77],"non-linear":[78],"activations":[79],"between":[80],"them.":[81],"In":[82],"addition,":[83],"Quasi-Orthogonal":[84],"Regularization":[85],"is":[86],"proposed":[87],"maintain":[89],"added":[91],"filters":[95],"as":[96],"orthogonal":[97],"matrixes.":[98],"After":[99],"training,":[100],"those":[101],"two":[102],"layers":[106],"be":[108],"fused":[109],"into":[110],"changing":[118],"network":[121],"architecture,":[122],"leaving":[123],"no":[124],"extra":[125],"computations":[126],"at":[127],"inference,":[128],"i.e.":[129],"parameter/FLOPs-free.":[130]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}