{"id":"https://openalex.org/W7133207135","doi":"https://doi.org/10.52202/068431-2642","title":"Learning Partial Equivariances from Data","display_name":"Learning Partial Equivariances from Data","publication_year":2022,"publication_date":"2022-01-01","ids":{"openalex":"https://openalex.org/W7133207135","doi":"https://doi.org/10.52202/068431-2642"},"language":"en","primary_location":{"id":"doi:10.52202/068431-2642","doi":"https://doi.org/10.52202/068431-2642","is_oa":false,"landing_page_url":"https://doi.org/10.52202/068431-2642","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"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":"Advances in Neural Information Processing Systems 35","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2110.10211","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127860550","display_name":"David W. Romero","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David W. Romero","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5056968977","display_name":"Suhas Lohit","orcid":"https://orcid.org/0000-0002-0392-3818","observed_orcids":["https://orcid.org/0000-0002-0392-3818"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Suhas Lohit","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.2109,"has_fulltext":true,"cited_by_count":19,"citation_normalized_percentile":{"value":0.86907778,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"36466","last_page":"36478"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.7060999870300293,"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":0.7060999870300293,"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":0.10440000146627426,"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/T10320","display_name":"Neural Networks and Applications","score":0.03590000048279762,"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/equivariant-neural-networks","display_name":"equivariant neural networks","score":0.6200000047683716},{"id":"https://openalex.org/keywords/rotation-equivariance","display_name":"rotation equivariance","score":0.6129999756813049},{"id":"https://openalex.org/keywords/natural-image-classification","display_name":"natural image classification","score":0.6980000138282776}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7735999822616577},{"id":"https://openalex.org/C74050887","wikidata":"https://www.wikidata.org/wiki/Q848368","display_name":"Rotation (mathematics)","level":2,"score":0.6567000150680542},{"id":"https://openalex.org/C96469262","wikidata":"https://www.wikidata.org/wiki/Q1324364","display_name":"Homogeneous space","level":2,"score":0.5896000266075134},{"id":"https://openalex.org/C2781311116","wikidata":"https://www.wikidata.org/wiki/Q83306","display_name":"Group (periodic table)","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5299000144004822},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.39340001344680786},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.373199999332428},{"id":"https://openalex.org/C207570611","wikidata":"https://www.wikidata.org/wiki/Q1256500","display_name":"Rotation group SO","level":2,"score":0.35910001397132874},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.3506999909877777},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.3296999931335449},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.3228999972343445},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3158000111579895}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.52202/068431-2642","doi":"https://doi.org/10.52202/068431-2642","is_oa":false,"landing_page_url":"https://doi.org/10.52202/068431-2642","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"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":"Advances in Neural Information Processing Systems 35","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2110.10211","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2110.10211","pdf_url":"https://arxiv.org/pdf/2110.10211","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"doi:10.48550/arxiv.2110.10211","doi":"https://doi.org/10.48550/arxiv.2110.10211","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2110.10211","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2110.10211","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2110.10211","pdf_url":"https://arxiv.org/pdf/2110.10211","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1250097850","display_name":null,"funder_award_id":"P16-25","funder_id":"https://openalex.org/F4320321800","funder_display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek"}],"funders":[{"id":"https://openalex.org/F4320315072","display_name":"Mitsubishi Electric Research Laboratories","ror":"https://ror.org/00e2e3n51"},{"id":"https://openalex.org/F4320321800","display_name":"Nederlandse Organisatie voor Wetenschappelijk Onderzoek","ror":"https://ror.org/04jsz6e67"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7133207135.pdf","grobid_xml":"https://content.openalex.org/works/W7133207135.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W1959608418","https://openalex.org/W2547875792","https://openalex.org/W2548228487","https://openalex.org/W2825179178","https://openalex.org/W2951970475","https://openalex.org/W3169403647","https://openalex.org/W3214507387","https://openalex.org/W4300537282","https://openalex.org/W4391156722"],"related_works":["https://openalex.org/W3207879461","https://openalex.org/W4400434628","https://openalex.org/W2279221249","https://openalex.org/W2951651290","https://openalex.org/W3168407340","https://openalex.org/W2751412330","https://openalex.org/W7133212273","https://openalex.org/W4312526534","https://openalex.org/W7006615242","https://openalex.org/W3213881022"],"abstract_inverted_index":{"Group":[0],"Convolutional":[1],"Neural":[2],"Networks":[3],"(G-CNNs)":[4],"constrain":[5],"learned":[6],"features":[7],"to":[8,18,37,83,105,119,124,148,157],"respect":[9],"the":[10,13,26,32,85,125,130],"symmetries":[11,23],"in":[12,25,47,68,109],"selected":[14],"group,":[15],"and":[16,41,95,154,161,210],"lead":[17],"better":[19,51,81,117],"generalization":[20],"when":[21,173,205],"these":[22],"appear":[24],"data.":[27,86],"If":[28],"this":[29,140],"is":[30,80,103,116,208],"not":[31],"case,":[33],"however,":[34],"equivariance":[35,78,102,115,134,156,172,207],"leads":[36],"overly":[38],"constrained":[39],"models":[40],"worse":[42],"performance.":[43],"Frequently,":[44],"transformations":[45,90],"occurring":[46],"data":[48],"can":[49],"be":[50],"represented":[52],"by":[53,60],"a":[54,57,61,64,74,110],"subset":[55],"of":[56,133,152,166,189],"group":[58,62],"than":[59],"as":[63,164],"whole,":[65],"e.g.,":[66,175,186],"rotations":[67],"$[-90^{\\circ},":[69],"90^{\\circ}]$.":[70],"In":[71,87,127,139],"such":[72],"cases,":[73],"model":[75],"that":[76,200],"respects":[77],"$\\textit{partially}$":[79],"suited":[82,118],"represent":[84],"addition,":[88],"relevant":[89],"may":[91,135],"differ":[92,136],"for":[93,176,187],"low":[94],"high-level":[96],"features.":[97],"For":[98],"instance,":[99],"full":[100,155,171,206],"rotation":[101,114],"useful":[104],"describe":[106,120],"edge":[107],"orientations":[108],"face,":[111],"but":[112,179],"partial":[113,153,201],"face":[121],"poses":[122],"relative":[123],"camera.":[126],"other":[128],"words,":[129],"optimal":[131],"level":[132],"per":[137],"layer.":[138],"work,":[141],"we":[142],"introduce":[143],"$\\textit{Partial":[144],"G-CNNs}$:":[145],"G-CNNs":[146,169,202,204],"able":[147],"learn":[149],"layer-wise":[150],"levels":[151],"discrete,":[158],"continuous":[159],"groups":[160],"combinations":[162],"thereof":[163],"part":[165],"training.":[167],"Partial":[168],"retain":[170],"beneficial,":[174],"rotated":[177],"MNIST,":[178],"adjust":[180],"it":[181,183],"whenever":[182],"becomes":[184],"harmful,":[185],"classification":[188],"6":[190],"/":[191],"9":[192],"digits":[193],"or":[194],"natural":[195],"images.":[196],"We":[197],"empirically":[198],"show":[199],"pair":[203],"advantageous,":[209],"outperform":[211],"them":[212],"otherwise.":[213]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":3},{"year":2024,"cited_by_count":3},{"year":2023,"cited_by_count":5},{"year":2022,"cited_by_count":4}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-03-03T00:00:00"}