{"id":"https://openalex.org/W7162673038","doi":"https://doi.org/10.48550/arxiv.2605.28578","title":"A Generalized Tikhonov Layer for Interpretable-by-design Graph Neural Networks","display_name":"A Generalized Tikhonov Layer for Interpretable-by-design Graph Neural Networks","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W7162673038","doi":"https://doi.org/10.48550/arxiv.2605.28578"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.28578","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28578","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.28578","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137205896","display_name":"Nicolas Tremblay","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tremblay, Nicolas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078878841","display_name":"Benjamin Ricaud","orcid":"https://orcid.org/0000-0002-7611-4448"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ricaud, Benjamin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5036254733","display_name":"Filippo Maria Bianchi","orcid":"https://orcid.org/0000-0002-7145-3846"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bianchi, Filippo Maria","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"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":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":0.9487000107765198,"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":0.9487000107765198,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.037300001829862595,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.0027000000700354576,"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/tikhonov-regularization","display_name":"Tikhonov regularization","score":0.7804999947547913},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5170000195503235},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.43540000915527344},{"id":"https://openalex.org/keywords/topology","display_name":"Topology (electrical circuits)","score":0.3928999900817871},{"id":"https://openalex.org/keywords/matrix","display_name":"Matrix (chemical analysis)","score":0.37119999527931213},{"id":"https://openalex.org/keywords/diagonal","display_name":"Diagonal","score":0.33820000290870667},{"id":"https://openalex.org/keywords/block-matrix","display_name":"Block matrix","score":0.3357999920845032},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3237999975681305}],"concepts":[{"id":"https://openalex.org/C152442038","wikidata":"https://www.wikidata.org/wiki/Q2778212","display_name":"Tikhonov regularization","level":3,"score":0.7804999947547913},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5530999898910522},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5170000195503235},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4648999869823456},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.43540000915527344},{"id":"https://openalex.org/C184720557","wikidata":"https://www.wikidata.org/wiki/Q7825049","display_name":"Topology (electrical circuits)","level":2,"score":0.3928999900817871},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3767000138759613},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C130367717","wikidata":"https://www.wikidata.org/wiki/Q189791","display_name":"Diagonal","level":2,"score":0.33820000290870667},{"id":"https://openalex.org/C85817219","wikidata":"https://www.wikidata.org/wiki/Q884772","display_name":"Block matrix","level":3,"score":0.3357999920845032},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.32989999651908875},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3237999975681305},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.31119999289512634},{"id":"https://openalex.org/C115178988","wikidata":"https://www.wikidata.org/wiki/Q772067","display_name":"Laplacian matrix","level":3,"score":0.3059999942779541},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.30300000309944153},{"id":"https://openalex.org/C203776342","wikidata":"https://www.wikidata.org/wiki/Q1378376","display_name":"Line graph","level":3,"score":0.2921999990940094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2831000089645386},{"id":"https://openalex.org/C61665672","wikidata":"https://www.wikidata.org/wiki/Q303100","display_name":"Graph isomorphism","level":4,"score":0.28220000863075256},{"id":"https://openalex.org/C180356752","wikidata":"https://www.wikidata.org/wiki/Q727035","display_name":"Adjacency matrix","level":3,"score":0.2671999931335449},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.25940001010894775},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C311688","wikidata":"https://www.wikidata.org/wiki/Q2393193","display_name":"Time complexity","level":2,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.28578","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28578","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.28578","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28578","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.6539679765701294,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,169],"propose":[1],"the":[2,29,38,43,52,77,82,119,127,131,159,163,181,184,187,212],"Tikhonov":[3,89,164,188],"layer,":[4,194],"a":[5,59,69,86,98,107,140,152,192,195],"graph":[6,30,54,88,129,207],"neural":[7],"network":[8],"layer":[9,78,165,189],"that":[10,91,113,157,171,211],"is":[11,51,81,143,146],"interpretable":[12,222],"by":[13,148],"design:":[14],"once":[15],"trained,":[16],"its":[17],"learned":[18,103],"parameters":[19],"directly":[20],"reveal":[21],"which":[22,26],"node":[23,114],"features":[24,117],"and":[25,67,202,223],"aspects":[27],"of":[28,63,85,133],"topology":[31],"were":[32],"leveraged":[33],"for":[34],"prediction.":[35],"In":[36],"practice,":[37],"layer's":[39],"propagation":[40,177],"matrix":[41,62],"takes":[42],"closed-form":[44],"$R":[45],"=":[46,57],"(p(L)+Q)^{-1}":[47],"Q$,":[48],"where":[49],"$L$":[50],"normalized":[53],"Laplacian,":[55],"$Q":[56],"diag(q_1,...,q_n)$":[58],"learnable":[60,70],"diagonal":[61],"positive":[64],"node-importance":[65,173],"scores,":[66,161],"$p(\\cdot)$":[68],"polynomial.":[71],"For":[72],"any":[73],"input":[74],"feature":[75],"$x$,":[76],"output":[79],"$Rx$":[80],"exact":[83],"minimizer":[84],"generalized":[87],"problem":[90],"trades":[92],"off":[93],"node-level":[94],"data":[95],"fidelity":[96],"against":[97],"topology-driven":[99],"regularization":[100],"penalty.":[101],"The":[102],"pair":[104],"$\\{\\{q_i\\},p\\}$":[105],"constitutes":[106],"built-in":[108],"explanation:":[109],"large":[110],"$q_i$":[111,123],"indicates":[112],"$i$'s":[115],"own":[116],"drive":[118],"prediction,":[120],"while":[121,162,220],"small":[122],"signals":[124],"reliance":[125],"on":[126,205],"local":[128],"topology;":[130],"shape":[132],"$p$":[134],"reveals":[135],"whether":[136],"homophily,":[137],"heterophily,":[138],"or":[139],"band-pass":[141],"response":[142],"exploited.":[144],"Expressivity":[145],"preserved":[147],"routing":[149],"complexity":[150],"through":[151],"dedicated,":[153],"arbitrarily":[154],"deep":[155],"Q-network":[156],"produces":[158],"importance":[160],"itself":[166],"remains":[167],"transparent.":[168],"prove":[170],"distinct":[172,176],"matrices":[174],"yield":[175],"operators,":[178],"structurally":[179],"coupling":[180],"explanation":[182],"to":[183],"computation.":[185],"Additionally,":[186],"provides,":[190],"in":[191],"single":[193],"global":[196],"receptive":[197],"field,":[198],"mitigating":[199],"both":[200],"oversmoothing":[201],"oversquashing.":[203],"Experiments":[204],"standard":[206],"classification":[208],"benchmarks":[209],"confirm":[210],"model":[213],"matches":[214],"(and":[215],"sometimes":[216],"outperforms)":[217],"opaque":[218],"baselines":[219],"producing":[221],"faithful":[224],"explanations.":[225]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-29T00:00:00"}