{"id":"https://openalex.org/W4378942300","doi":"https://doi.org/10.48550/arxiv.2305.18378","title":"Disentanglement via Latent Quantization","display_name":"Disentanglement via Latent Quantization","publication_year":2023,"publication_date":"2023-05-28","ids":{"openalex":"https://openalex.org/W4378942300","doi":"https://doi.org/10.48550/arxiv.2305.18378"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2305.18378","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.18378","pdf_url":"https://arxiv.org/pdf/2305.18378","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2305.18378","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055442462","display_name":"Kyle Hsu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hsu, Kyle","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013128519","display_name":"William Dorrell","orcid":"https://orcid.org/0000-0002-6748-3401"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dorrell, Will","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036397805","display_name":"James C. R. Whittington","orcid":"https://orcid.org/0000-0001-5680-5586"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Whittington, James C. R.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100621605","display_name":"Jiajun Wu","orcid":"https://orcid.org/0000-0002-4176-343X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Jiajun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5005431772","display_name":"Chelsea Finn","orcid":"https://orcid.org/0000-0001-6298-0874"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Finn, Chelsea","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9962999820709229,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9962999820709229,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.9958000183105469,"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/T11512","display_name":"Anomaly Detection Techniques and Applications","score":0.9739999771118164,"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/autoencoder","display_name":"Autoencoder","score":0.772046685218811},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6026239991188049},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.5213733315467834},{"id":"https://openalex.org/keywords/regularization","display_name":"Regularization (linguistics)","score":0.4805926978588104},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4605070948600769},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.445487380027771},{"id":"https://openalex.org/keywords/vector-quantization","display_name":"Vector quantization","score":0.4273570477962494},{"id":"https://openalex.org/keywords/probabilistic-latent-semantic-analysis","display_name":"Probabilistic latent semantic analysis","score":0.41101136803627014},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.39991098642349243},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.38078123331069946},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.34726232290267944},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.21294769644737244}],"concepts":[{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.772046685218811},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6026239991188049},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.5213733315467834},{"id":"https://openalex.org/C2776135515","wikidata":"https://www.wikidata.org/wiki/Q17143721","display_name":"Regularization (linguistics)","level":2,"score":0.4805926978588104},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4605070948600769},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.445487380027771},{"id":"https://openalex.org/C199833920","wikidata":"https://www.wikidata.org/wiki/Q612536","display_name":"Vector quantization","level":2,"score":0.4273570477962494},{"id":"https://openalex.org/C112933361","wikidata":"https://www.wikidata.org/wiki/Q2845258","display_name":"Probabilistic latent semantic analysis","level":2,"score":0.41101136803627014},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39991098642349243},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.38078123331069946},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.34726232290267944},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.21294769644737244}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2305.18378","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.18378","pdf_url":"https://arxiv.org/pdf/2305.18378","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2305.18378","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2305.18378","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":"pmh:oai:arXiv.org:2305.18378","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2305.18378","pdf_url":"https://arxiv.org/pdf/2305.18378","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.47999998927116394}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2988134182","https://openalex.org/W2884410131","https://openalex.org/W1603253275","https://openalex.org/W120501756","https://openalex.org/W111011176","https://openalex.org/W2763292376","https://openalex.org/W2138996412","https://openalex.org/W2951183560","https://openalex.org/W2097596242","https://openalex.org/W2906932471"],"abstract_inverted_index":{"In":[0,46,214],"disentangled":[1],"representation":[2],"learning,":[3],"a":[4,11,40,79,111,126,172,208],"model":[5,26,90,138],"is":[6,27,180],"asked":[7],"to":[8,56,106,124,129,135,154],"tease":[9],"apart":[10],"dataset's":[12],"underlying":[13],"sources":[14],"of":[15,21,114,148,175,204,211],"variation":[16],"and":[17,57,86,160,186,202],"represent":[18],"them":[19],"independently":[20],"one":[22],"another.":[23],"Since":[24],"the":[25,71,99,104,122,137,145,200],"provided":[28],"with":[29,78,194],"no":[30],"ground":[31],"truth":[32],"information":[33,184],"about":[34],"these":[35,228],"sources,":[36],"inductive":[37,52],"biases":[38],"take":[39],"paramount":[41],"role":[42],"in":[43,119,183,190,227],"enabling":[44],"disentanglement.":[45],"this":[47,67,140,149],"work,":[48],"we":[49,65,168],"construct":[50,108],"an":[51,60,93],"bias":[53],"towards":[54,139],"encoding":[55],"decoding":[58],"from":[59,110,224],"organized":[61],"latent":[62,72,100,196],"space.":[63],"Concretely,":[64],"do":[66],"by":[68,151],"(i)":[69],"quantizing":[70],"space":[73,101],"into":[74],"discrete":[75],"code":[76],"vectors":[77],"separate":[80],"learnable":[81],"scalar":[82,116],"codebook":[83],"per":[84],"dimension":[85],"(ii)":[87],"applying":[88],"strong":[89,222],"regularization":[91],"via":[92],"unusually":[94],"high":[95],"weight":[96],"decay.":[97],"Intuitively,":[98],"design":[102],"forces":[103],"encoder":[105],"combinatorially":[107],"codes":[109],"small":[112],"number":[113],"distinct":[115],"values,":[117],"which":[118],"turn":[120],"enables":[121],"decoder":[123],"assign":[125],"consistent":[127],"meaning":[128],"each":[130],"value.":[131],"Regularization":[132],"then":[133],"serves":[134],"drive":[136],"parsimonious":[141],"strategy.":[142],"We":[143],"demonstrate":[144],"broad":[146],"applicability":[147],"approach":[150],"adding":[152],"it":[153],"both":[155],"basic":[156],"data-reconstructing":[157],"(vanilla":[158],"autoencoder)":[159],"latent-reconstructing":[161],"(InfoGAN)":[162],"generative":[163],"models.":[164],"For":[165],"reliable":[166],"evaluation,":[167],"also":[169],"propose":[170],"InfoMEC,":[171],"new":[173],"set":[174],"metrics":[176],"for":[177],"disentanglement":[178,230],"that":[179],"cohesively":[181],"grounded":[182],"theory":[185],"fixes":[187],"well-established":[188],"shortcomings":[189],"previous":[191],"metrics.":[192],"Together":[193],"regularization,":[195],"quantization":[197],"dramatically":[198],"improves":[199],"modularity":[201],"explicitness":[203],"learned":[205],"representations":[206],"on":[207],"representative":[209],"suite":[210],"benchmark":[212],"datasets.":[213],"particular,":[215],"our":[216],"quantized-latent":[217],"autoencoder":[218],"(QLAE)":[219],"consistently":[220],"outperforms":[221],"methods":[223],"prior":[225],"work":[226],"key":[229],"properties":[231],"without":[232],"compromising":[233],"data":[234],"reconstruction.":[235]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}