{"id":"https://openalex.org/W2791344998","doi":"https://doi.org/10.1007/978-3-030-01249-6_44","title":"A Dataset and Architecture for Visual Reasoning with a Working Memory","display_name":"A Dataset and Architecture for Visual Reasoning with a Working Memory","publication_year":2018,"publication_date":"2018-01-01","ids":{"openalex":"https://openalex.org/W2791344998","doi":"https://doi.org/10.1007/978-3-030-01249-6_44","mag":"2791344998"},"language":"en","primary_location":{"id":"doi:10.1007/978-3-030-01249-6_44","is_oa":false,"landing_page_url":"https://doi.org/10.1007/978-3-030-01249-6_44","pdf_url":null,"source":{"id":"https://openalex.org/S106296714","display_name":"Lecture notes in computer science","issn_l":"0302-9743","issn":["0302-9743","1611-3349"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"book series"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Lecture Notes in Computer Science","raw_type":"book-chapter"},"type":"preprint","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/1803.06092","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5053839400","display_name":"Guangyu Robert Yang","orcid":"https://orcid.org/0000-0002-8919-4248"},"institutions":[{"id":"https://openalex.org/I78577930","display_name":"Columbia University","ror":"https://ror.org/00hj8s172","country_code":"US","type":"education","lineage":["https://openalex.org/I78577930"]},{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Guangyu Robert Yang","raw_affiliation_strings":["Columbia University","New York University"],"affiliations":[{"raw_affiliation_string":"Columbia University","institution_ids":["https://openalex.org/I78577930"]},{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030915482","display_name":"Igor Ganichev","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Igor Ganichev","raw_affiliation_strings":["Google","[Google Brain]"],"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Google Brain]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101509100","display_name":"Xiao\u2010Jing Wang","orcid":"https://orcid.org/0000-0003-3124-8474"},"institutions":[{"id":"https://openalex.org/I102475099","display_name":"Sup\u00e9lec","ror":"https://ror.org/00n7gwn90","country_code":"FR","type":"education","lineage":["https://openalex.org/I102475099"]},{"id":"https://openalex.org/I15196421","display_name":"University of Applied Sciences and Arts of Southern Switzerland","ror":"https://ror.org/05ep8g269","country_code":"CH","type":"education","lineage":["https://openalex.org/I15196421"]},{"id":"https://openalex.org/I57206974","display_name":"New York University","ror":"https://ror.org/0190ak572","country_code":"US","type":"education","lineage":["https://openalex.org/I57206974"]},{"id":"https://openalex.org/I4210102264","display_name":"Shandong University of Political Science and Law","ror":"https://ror.org/01b2j5886","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210102264"]}],"countries":["CH","CN","FR","US"],"is_corresponding":false,"raw_author_name":"Xiao-Jing Wang","raw_affiliation_strings":["New York University","New York University<sup>**</sup>"],"affiliations":[{"raw_affiliation_string":"New York University","institution_ids":["https://openalex.org/I57206974"]},{"raw_affiliation_string":"New York University<sup>**</sup>","institution_ids":["https://openalex.org/I15196421","https://openalex.org/I4210102264","https://openalex.org/I102475099"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032356827","display_name":"Jonathon Shlens","orcid":"https://orcid.org/0000-0001-9513-4244"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jonathon Shlens","raw_affiliation_strings":["Google","[Google Brain]"],"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Google Brain]","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5054372964","display_name":"David Sussillo","orcid":"https://orcid.org/0000-0003-1620-1264"},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Sussillo","raw_affiliation_strings":["Google","[Google Brain]"],"affiliations":[{"raw_affiliation_string":"Google","institution_ids":["https://openalex.org/I1291425158"]},{"raw_affiliation_string":"[Google Brain]","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":4,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5053839400"],"corresponding_institution_ids":["https://openalex.org/I57206974","https://openalex.org/I78577930"],"apc_list":{"value":5000,"currency":"EUR","value_usd":5392},"apc_paid":null,"fwci":1.1685,"has_fulltext":true,"cited_by_count":13,"citation_normalized_percentile":{"value":0.83441774,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"729","last_page":"745"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9983000159263611,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9950000047683716,"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/computer-science","display_name":"Computer science","score":0.7532375454902649},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6523975729942322},{"id":"https://openalex.org/keywords/cog","display_name":"Cog","score":0.6465251445770264},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.6460845470428467},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6001640558242798},{"id":"https://openalex.org/keywords/cognitive-architecture","display_name":"Cognitive architecture","score":0.560468316078186},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.5142056941986084},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.48853787779808044},{"id":"https://openalex.org/keywords/cognition","display_name":"Cognition","score":0.4319896697998047},{"id":"https://openalex.org/keywords/cognitive-science","display_name":"Cognitive science","score":0.3662795424461365},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3655053377151489},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.12029364705085754}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7532375454902649},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6523975729942322},{"id":"https://openalex.org/C4661277","wikidata":"https://www.wikidata.org/wiki/Q1054157","display_name":"Cog","level":2,"score":0.6465251445770264},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.6460845470428467},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6001640558242798},{"id":"https://openalex.org/C20854674","wikidata":"https://www.wikidata.org/wiki/Q4386060","display_name":"Cognitive architecture","level":3,"score":0.560468316078186},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.5142056941986084},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.48853787779808044},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.4319896697998047},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.3662795424461365},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3655053377151489},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.12029364705085754},{"id":"https://openalex.org/C169760540","wikidata":"https://www.wikidata.org/wiki/Q207011","display_name":"Neuroscience","level":1,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0},{"id":"https://openalex.org/C187736073","wikidata":"https://www.wikidata.org/wiki/Q2920921","display_name":"Management","level":1,"score":0.0},{"id":"https://openalex.org/C153349607","wikidata":"https://www.wikidata.org/wiki/Q36649","display_name":"Visual arts","level":1,"score":0.0},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.0}],"mesh":[],"locations_count":5,"locations":[{"id":"doi:10.1007/978-3-030-01249-6_44","is_oa":false,"landing_page_url":"https://doi.org/10.1007/978-3-030-01249-6_44","pdf_url":null,"source":{"id":"https://openalex.org/S106296714","display_name":"Lecture notes in computer science","issn_l":"0302-9743","issn":["0302-9743","1611-3349"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"book series"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Lecture Notes in Computer Science","raw_type":"book-chapter"},{"id":"doi:10.32470/ccn.2018.1075-0","is_oa":false,"landing_page_url":"https://doi.org/10.32470/ccn.2018.1075-0","pdf_url":null,"source":{"id":"https://openalex.org/S4306498296","display_name":"2018 Conference on Cognitive Computational Neuroscience","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":"2018 Conference on Cognitive Computational Neuroscience","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:1803.06092","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1803.06092","pdf_url":"https://arxiv.org/pdf/1803.06092","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"mag:2791344998","is_oa":true,"landing_page_url":"https://arxiv.org/pdf/1803.06092","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"arXiv (Cornell University)","raw_type":null},{"id":"doi:10.48550/arxiv.1803.06092","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.1803.06092","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:1803.06092","is_oa":true,"landing_page_url":"http://arxiv.org/abs/1803.06092","pdf_url":"https://arxiv.org/pdf/1803.06092","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.5699999928474426}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2791344998.pdf","grobid_xml":"https://content.openalex.org/works/W2791344998.grobid-xml"},"referenced_works_count":56,"referenced_works":["https://openalex.org/W569478347","https://openalex.org/W1514535095","https://openalex.org/W1522301498","https://openalex.org/W1525961042","https://openalex.org/W1747565446","https://openalex.org/W1757796397","https://openalex.org/W1898975967","https://openalex.org/W1907121963","https://openalex.org/W1927052826","https://openalex.org/W1933349210","https://openalex.org/W1996418862","https://openalex.org/W2000697043","https://openalex.org/W2005814556","https://openalex.org/W2016053056","https://openalex.org/W2033708181","https://openalex.org/W2039845328","https://openalex.org/W2047125104","https://openalex.org/W2056941131","https://openalex.org/W2064675550","https://openalex.org/W2074446477","https://openalex.org/W2074619556","https://openalex.org/W2076217198","https://openalex.org/W2123086804","https://openalex.org/W2124517301","https://openalex.org/W2128469230","https://openalex.org/W2131774270","https://openalex.org/W2145339207","https://openalex.org/W2151137320","https://openalex.org/W2325237720","https://openalex.org/W2463267937","https://openalex.org/W2524365899","https://openalex.org/W2530887700","https://openalex.org/W2545656684","https://openalex.org/W2561715562","https://openalex.org/W2606220156","https://openalex.org/W2613404084","https://openalex.org/W2613526370","https://openalex.org/W2619947201","https://openalex.org/W2624614404","https://openalex.org/W2738724892","https://openalex.org/W2749807327","https://openalex.org/W2752440022","https://openalex.org/W2760103357","https://openalex.org/W2789500754","https://openalex.org/W2949117887","https://openalex.org/W2949626814","https://openalex.org/W2950761309","https://openalex.org/W2962716332","https://openalex.org/W2962749469","https://openalex.org/W2963224792","https://openalex.org/W2963260436","https://openalex.org/W2963656855","https://openalex.org/W2964308564","https://openalex.org/W4211081781","https://openalex.org/W4213346165","https://openalex.org/W6743198006"],"related_works":["https://openalex.org/W2561715562","https://openalex.org/W2963921132","https://openalex.org/W2950761309","https://openalex.org/W2962716332","https://openalex.org/W2194775991","https://openalex.org/W2997399961","https://openalex.org/W2964118342","https://openalex.org/W2786209943","https://openalex.org/W2753474648","https://openalex.org/W3034205192","https://openalex.org/W3039442841","https://openalex.org/W2903184186","https://openalex.org/W3129814727","https://openalex.org/W2962781483","https://openalex.org/W2982389186","https://openalex.org/W2339712187","https://openalex.org/W3008526508","https://openalex.org/W2794284373","https://openalex.org/W2963880187","https://openalex.org/W3016173130"],"abstract_inverted_index":{"A":[0],"vexing":[1],"problem":[2,65],"in":[3,12,19,35,54,126,162],"artificial":[4],"intelligence":[5],"is":[6,59],"reasoning":[7,32,81],"about":[8],"events":[9],"that":[10,86,101,128,156],"occur":[11],"complex,":[13],"changing":[14],"visual":[15,31,45,78],"stimuli":[16],"such":[17],"as":[18,111,113],"video":[20,67],"analysis":[21],"or":[22],"game":[23],"play.":[24],"Inspired":[25],"by":[26],"a":[27,97,163],"rich":[28],"tradition":[29],"of":[30,66,73,116,123,148],"and":[33,38,47,56,79,82],"memory":[34,83],"cognitive":[36],"psychology":[37],"neuroscience,":[39],"we":[40],"developed":[41],"an":[42],"artificial,":[43],"configurable":[44],"question":[46],"answer":[48],"dataset":[49],"(COG)":[50],"to":[51,77,133,142,166],"parallel":[52],"experiments":[53],"humans":[55],"animals.":[57],"COG":[58,118,124,154],"much":[60],"simpler":[61],"than":[62],"the":[63,74,117,137,149,157,160],"general":[64],"analysis,":[68],"yet":[69],"it":[70],"addresses":[71],"many":[72,143],"problems":[75,85],"relating":[76],"logical":[80],"\u2013":[84],"remain":[87],"challenging":[88,132],"for":[89],"modern":[90],"deep":[91,98],"learning":[92,99],"architectures.":[93],"We":[94],"additionally":[95],"propose":[96],"architecture":[100],"performs":[102],"competitively":[103],"on":[104,153],"other":[105],"diagnostic":[106],"VQA":[107],"datasets":[108,127],"(i.e.":[109],"CLEVR)":[110],"well":[112],"easy":[114],"settings":[115,122],"dataset.":[119],"However,":[120],"several":[121],"result":[125],"are":[129],"progressively":[130],"more":[131],"learn.":[134],"After":[135],"training,":[136],"network":[138,150,158],"can":[139],"zero-shot":[140],"generalize":[141],"new":[144],"tasks.":[145],"Preliminary":[146],"analyses":[147],"architectures":[151],"trained":[152],"demonstrate":[155],"accomplishes":[159],"task":[161],"manner":[164],"interpretable":[165],"humans.":[167]},"counts_by_year":[{"year":2022,"cited_by_count":2},{"year":2021,"cited_by_count":3},{"year":2020,"cited_by_count":2},{"year":2019,"cited_by_count":3},{"year":2018,"cited_by_count":3}],"updated_date":"2026-03-10T16:38:18.471706","created_date":"2025-10-10T00:00:00"}