{"id":"https://openalex.org/W4366851092","doi":"https://doi.org/10.48550/arxiv.2304.10851","title":"What Do GNNs Actually Learn? Towards Understanding their Representations","display_name":"What Do GNNs Actually Learn? Towards Understanding their Representations","publication_year":2023,"publication_date":"2023-04-21","ids":{"openalex":"https://openalex.org/W4366851092","doi":"https://doi.org/10.48550/arxiv.2304.10851"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2304.10851","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.10851","pdf_url":"https://arxiv.org/pdf/2304.10851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2304.10851","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5011742954","display_name":"Giannis Nikolentzos","orcid":"https://orcid.org/0000-0002-0336-5879"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Nikolentzos, Giannis","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072113617","display_name":"Michail Chatzianastasis","orcid":"https://orcid.org/0000-0002-9905-1646"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chatzianastasis, Michail","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5085652160","display_name":"Michalis Vazirgiannis","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vazirgiannis, Michalis","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5011742954"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9997000098228455,"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"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9997000098228455,"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/T12292","display_name":"Graph Theory and Algorithms","score":0.9521999955177307,"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/T10203","display_name":"Recommender Systems and Techniques","score":0.9341999888420105,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6838563084602356},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5921367406845093},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.5751107931137085},{"id":"https://openalex.org/keywords/node","display_name":"Node (physics)","score":0.5724684000015259},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5410526990890503},{"id":"https://openalex.org/keywords/feature-learning","display_name":"Feature learning","score":0.5285968780517578},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4261396825313568},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.41427066922187805},{"id":"https://openalex.org/keywords/graph-isomorphism","display_name":"Graph isomorphism","score":0.4137420058250427},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.32347404956817627},{"id":"https://openalex.org/keywords/line-graph","display_name":"Line graph","score":0.14500680565834045}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6838563084602356},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5921367406845093},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.5751107931137085},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.5724684000015259},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5410526990890503},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.5285968780517578},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4261396825313568},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41427066922187805},{"id":"https://openalex.org/C61665672","wikidata":"https://www.wikidata.org/wiki/Q303100","display_name":"Graph isomorphism","level":4,"score":0.4137420058250427},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32347404956817627},{"id":"https://openalex.org/C203776342","wikidata":"https://www.wikidata.org/wiki/Q1378376","display_name":"Line graph","level":3,"score":0.14500680565834045},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2304.10851","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.10851","pdf_url":"https://arxiv.org/pdf/2304.10851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2304.10851","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2304.10851","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2304.10851","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2304.10851","pdf_url":"https://arxiv.org/pdf/2304.10851","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4366851092.pdf","grobid_xml":"https://content.openalex.org/works/W4366851092.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2991205486","https://openalex.org/W2073067110","https://openalex.org/W2905073546","https://openalex.org/W2905271011","https://openalex.org/W2793270624","https://openalex.org/W3164948662","https://openalex.org/W4289536128","https://openalex.org/W3153597579","https://openalex.org/W4298151006","https://openalex.org/W4285218279"],"abstract_inverted_index":{"In":[0,59],"recent":[1],"years,":[2],"graph":[3,15],"neural":[4],"networks":[5],"(GNNs)":[6],"have":[7],"achieved":[8],"great":[9],"success":[10],"in":[11,148,175,192],"the":[12,25,50,68,89,108,120,128,134,141,149,152,156,160,166,186],"field":[13],"of":[14,27,36,100,102,122,130,144,159,169,188],"representation":[16],"learning.":[17],"Although":[18],"prior":[19],"work":[20],"has":[21],"shed":[22],"light":[23],"on":[24,133,201],"expressiveness":[26],"those":[28,57],"models":[29,81,94,116,161],"(\\ie":[30],"whether":[31],"they":[32],"can":[33,172],"distinguish":[34],"pairs":[35],"non-isomorphic":[37],"graphs),":[38],"it":[39],"is":[40,47,197],"still":[41],"not":[42],"clear":[43],"what":[44],"structural":[45],"information":[46],"encoded":[48],"into":[49],"node":[51,69,131],"representations":[52,70,84,90,143,153],"that":[53,79,105,139,171,190],"are":[54,95,162],"learned":[55,71,91,135,154],"by":[56,66,72,92],"models.":[58,76],"this":[60,64],"paper,":[61],"we":[62,126],"address":[63],"gap":[65],"studying":[67],"four":[73],"standard":[74],"GNN":[75],"We":[77,110,137,179],"find":[78,138],"some":[80,98],"produce":[82],"identical":[83],"for":[85,114],"all":[86,145],"nodes,":[87],"while":[88],"other":[93],"linked":[96],"to":[97,119,165,184],"notion":[99],"walks":[101],"specific":[103],"length":[104],"start":[106],"from":[107],"nodes.":[109],"establish":[111],"Lipschitz":[112],"bounds":[113],"these":[115],"with":[117],"respect":[118],"number":[121],"(normalized)":[123],"walks.":[124],"Additionally,":[125],"investigate":[127],"influence":[129],"features":[132,168],"representations.":[136],"if":[140],"initial":[142,167],"nodes":[146,170],"point":[147],"same":[150],"direction,":[151],"at":[155],"$k$-th":[157],"layer":[158],"also":[163,180],"related":[164],"be":[173],"reached":[174],"exactly":[176],"$k$":[177],"steps.":[178],"apply":[181],"our":[182],"findings":[183],"understand":[185],"phenomenon":[187],"oversquashing":[189],"occurs":[191],"GNNs.":[193],"Our":[194],"theoretical":[195],"analysis":[196],"validated":[198],"through":[199],"experiments":[200],"synthetic":[202],"and":[203],"real-world":[204],"datasets.":[205]},"counts_by_year":[],"updated_date":"2026-03-09T08:58:05.943551","created_date":"2025-10-10T00:00:00"}