{"id":"https://openalex.org/W4302010995","doi":"https://doi.org/10.48550/arxiv.2210.00319","title":"PathFinder: Discovering Decision Pathways in Deep Neural Networks","display_name":"PathFinder: Discovering Decision Pathways in Deep Neural Networks","publication_year":2022,"publication_date":"2022-10-01","ids":{"openalex":"https://openalex.org/W4302010995","doi":"https://doi.org/10.48550/arxiv.2210.00319"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2210.00319","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2210.00319","pdf_url":"https://arxiv.org/pdf/2210.00319","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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2210.00319","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5088262368","display_name":"Ozan \u0130rsoy","orcid":"https://orcid.org/0000-0002-7123-8361","observed_orcids":["https://orcid.org/0000-0002-7123-8361"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"\u0130rsoy, Ozan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5066373250","display_name":"Ethem Alpayd\u0131n","orcid":"https://orcid.org/0000-0001-7506-0321","observed_orcids":["https://orcid.org/0000-0001-7506-0321"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Alpayd\u0131n, Ethem","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":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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":1.0,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":1.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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/T10320","display_name":"Neural Networks and Applications","score":0.0,"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/decision-paths","display_name":"decision paths","score":0.6579999923706055},{"id":"https://openalex.org/keywords/explainable-ai","display_name":"explainable AI","score":0.7139999866485596},{"id":"https://openalex.org/keywords/deep-neural-network","display_name":"deep neural network","score":0.781000018119812},{"id":"https://openalex.org/keywords/hidden-layer-activations","display_name":"hidden layer activations","score":0.640999972820282},{"id":"https://openalex.org/keywords/sankey-diagrams","display_name":"Sankey diagrams","score":0.7850000262260437},{"id":"https://openalex.org/keywords/outlier-detection","display_name":"outlier detection","score":0.8519999980926514},{"id":"https://openalex.org/keywords/recurrent-neural-networks","display_name":"recurrent neural networks","score":0.6759999990463257},{"id":"https://openalex.org/keywords/fully-connected-layers","display_name":"fully connected layers","score":0.625}],"concepts":[{"id":"https://openalex.org/C190502265","wikidata":"https://www.wikidata.org/wiki/Q17069496","display_name":"MNIST database","level":3,"score":0.9085671901702881},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7674291729927063},{"id":"https://openalex.org/C2778940482","wikidata":"https://www.wikidata.org/wiki/Q7144753","display_name":"Pathfinder","level":2,"score":0.6176319718360901},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.6075224280357361},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.5589651465415955},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5558821558952332},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.5427073240280151},{"id":"https://openalex.org/C164866538","wikidata":"https://www.wikidata.org/wiki/Q367351","display_name":"Cluster (spacecraft)","level":2,"score":0.5000896453857422},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.4735143482685089},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4410303831100464},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.373484343290329},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3363994061946869},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.08978614211082458},{"id":"https://openalex.org/C161191863","wikidata":"https://www.wikidata.org/wiki/Q199655","display_name":"Library science","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2210.00319","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2210.00319","pdf_url":"https://arxiv.org/pdf/2210.00319","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.2210.00319","doi":"https://doi.org/10.48550/arxiv.2210.00319","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2210.00319","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:2210.00319","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2210.00319","pdf_url":"https://arxiv.org/pdf/2210.00319","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":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415315617","https://openalex.org/W6925680399","https://openalex.org/W3206338095","https://openalex.org/W4390586901","https://openalex.org/W2964089344","https://openalex.org/W3011855871","https://openalex.org/W4210705368","https://openalex.org/W4409796137","https://openalex.org/W4382467091","https://openalex.org/W3199774549"],"abstract_inverted_index":{"Explainability":[0],"is":[1,17,33],"becoming":[2],"an":[3],"increasingly":[4],"important":[5],"topic":[6],"for":[7,62],"deep":[8],"neural":[9],"networks.":[10],"Though":[11],"the":[12,41,49,70,83,87,94,105,135],"operation":[13],"in":[14,24,30,48,53,72],"convolutional":[15],"layers":[16,51,74],"easier":[18],"to":[19,76,86,138],"understand,":[20],"processing":[21],"becomes":[22],"opaque":[23],"fully-connected":[25],"layers.":[26],"The":[27],"basic":[28],"idea":[29],"our":[31,54,144],"work":[32],"that":[34,91,127],"each":[35,63],"instance,":[36],"as":[37,79],"it":[38],"flows":[39,81],"through":[40],"layers,":[42,106],"causes":[43],"a":[44,98],"different":[45],"activation":[46,60,80],"pattern":[47],"hidden":[50,64],"and":[52,66,121,155,159],"Paths":[55],"methodology,":[56],"we":[57,108],"cluster":[58,102],"these":[59],"vectors":[61],"layer":[65,85],"then":[67],"see":[68],"how":[69,115],"clusters":[71],"successive":[73],"connect":[75],"one":[77,160],"another":[78],"from":[82],"input":[84],"output.":[88],"We":[89,131,142],"find":[90],"instances":[92],"of":[93,101],"same":[95],"class":[96],"follow":[97,128],"small":[99],"number":[100],"sequences":[103],"over":[104],"which":[107],"name":[109],"``decision":[110],"paths.\"":[111],"Such":[112],"paths":[113],"explain":[114],"classification":[116],"decisions":[117],"are":[118],"typically":[119],"made,":[120],"also":[122,132],"help":[123],"us":[124],"determine":[125],"outliers":[126],"unusual":[129],"paths.":[130],"propose":[133],"using":[134],"Sankey":[136],"diagram":[137],"visualize":[139],"such":[140],"pathways.":[141],"validate":[143],"method":[145],"with":[146],"experiments":[147],"on":[148,153,164],"two":[149],"feed-forward":[150],"networks":[151],"trained":[152,163],"MNIST":[154],"CELEB":[156],"data":[157],"sets,":[158],"recurrent":[161],"network":[162],"PenDigits.":[165]},"counts_by_year":[],"updated_date":"2026-08-26T07:47:46.906454","created_date":"2022-10-06T00:00:00"}