{"id":"https://openalex.org/W3207589413","doi":"https://doi.org/10.48550/arxiv.2110.11271","title":"Analyzing and Improving the Optimization Landscape of Noise-Contrastive Estimation","display_name":"Analyzing and Improving the Optimization Landscape of Noise-Contrastive Estimation","publication_year":2021,"publication_date":"2021-10-21","ids":{"openalex":"https://openalex.org/W3207589413","doi":"https://doi.org/10.48550/arxiv.2110.11271","mag":"3207589413"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2110.11271","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2110.11271","pdf_url":"https://arxiv.org/pdf/2110.11271","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/2110.11271","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5007869267","display_name":"Bingbin Liu","orcid":"https://orcid.org/0000-0002-6621-0356","observed_orcids":["https://orcid.org/0000-0002-6621-0356"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Liu, Bingbin","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075587107","display_name":"Elan Rosenfeld","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rosenfeld, Elan","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053209283","display_name":"Pradeep Ravikumar","orcid":"https://orcid.org/0000-0001-5635-5765","observed_orcids":["https://orcid.org/0000-0001-5635-5765"]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ravikumar, Pradeep","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5014972240","display_name":"Andrej Risteski","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Risteski, Andrej","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I74973139"],"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.38040000200271606,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.38040000200271606,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.13279999792575836,"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/T10320","display_name":"Neural Networks and Applications","score":0.11869999766349792,"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/noise-contrastive-estimation","display_name":"noise contrastive estimation","score":0.9990000128746033},{"id":"https://openalex.org/keywords/loss-landscape","display_name":"loss landscape","score":0.6629999876022339},{"id":"https://openalex.org/keywords/flat-minima","display_name":"flat minima","score":0.7879999876022339},{"id":"https://openalex.org/keywords/exponential-loss","display_name":"exponential loss","score":0.8379999995231628},{"id":"https://openalex.org/keywords/exponential-family","display_name":"exponential family","score":0.8859999775886536},{"id":"https://openalex.org/keywords/statistical-consistency","display_name":"statistical consistency","score":0.6729999780654907},{"id":"https://openalex.org/keywords/optimization-landscape","display_name":"optimization landscape","score":0.41200000047683716}],"concepts":[{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.7519508600234985},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6218937039375305},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.6148642897605896},{"id":"https://openalex.org/C151376022","wikidata":"https://www.wikidata.org/wiki/Q168698","display_name":"Exponential function","level":2,"score":0.5295794010162354},{"id":"https://openalex.org/C55974624","wikidata":"https://www.wikidata.org/wiki/Q1188504","display_name":"Exponential family","level":2,"score":0.506206214427948},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.46740204095840454},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.4672520160675049},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.4514068067073822},{"id":"https://openalex.org/C200378446","wikidata":"https://www.wikidata.org/wiki/Q4147391","display_name":"Gradient noise","level":5,"score":0.4290426969528198},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.41228240728378296},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4114059805870056},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.41016459465026855},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3415409028530121},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.330617219209671},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3167809247970581},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.3071885108947754},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.14157983660697937},{"id":"https://openalex.org/C187612029","wikidata":"https://www.wikidata.org/wiki/Q17083130","display_name":"Noise floor","level":4,"score":0.12419691681861877},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.11340746283531189},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.10408008098602295},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.06881198287010193},{"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}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2110.11271","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2110.11271","pdf_url":"https://arxiv.org/pdf/2110.11271","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.2110.11271","doi":"https://doi.org/10.48550/arxiv.2110.11271","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2110.11271","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:2110.11271","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2110.11271","pdf_url":"https://arxiv.org/pdf/2110.11271","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W82247633","https://openalex.org/W91088564","https://openalex.org/W1483912251","https://openalex.org/W1513873506","https://openalex.org/W1768488313","https://openalex.org/W2013164703","https://openalex.org/W2097732278","https://openalex.org/W2120340025","https://openalex.org/W2120861206","https://openalex.org/W2131939418","https://openalex.org/W2138204974","https://openalex.org/W2152790380","https://openalex.org/W2159611450","https://openalex.org/W2405787682","https://openalex.org/W2419501139","https://openalex.org/W2439299270","https://openalex.org/W2841543429","https://openalex.org/W2887997457","https://openalex.org/W2897723259","https://openalex.org/W2922772346","https://openalex.org/W2949979820","https://openalex.org/W2962755094","https://openalex.org/W2971899460","https://openalex.org/W2980360762","https://openalex.org/W2994434574","https://openalex.org/W2995024809","https://openalex.org/W3009318622","https://openalex.org/W3035058308","https://openalex.org/W3035166812","https://openalex.org/W3036122622","https://openalex.org/W3038081132","https://openalex.org/W3148140980","https://openalex.org/W3194546735","https://openalex.org/W4297808394"],"related_works":["https://openalex.org/W4399115952","https://openalex.org/W4302010541","https://openalex.org/W4392223077","https://openalex.org/W4403797585","https://openalex.org/W4380687275","https://openalex.org/W2938296692","https://openalex.org/W4415180120","https://openalex.org/W2803256244","https://openalex.org/W2888075180","https://openalex.org/W4221165035"],"abstract_inverted_index":{"Noise-contrastive":[0],"estimation":[1],"(NCE)":[2],"is":[3,25,43,79],"a":[4,52,101,131],"statistically":[5],"consistent":[6],"method":[7],"for":[8,27,70,113],"learning":[9],"unnormalized":[10],"probabilistic":[11],"models.":[12],"It":[13],"has":[14],"been":[15,35],"empirically":[16],"observed":[17],"that":[18],"the":[19,22,48,119,124],"choice":[20],"of":[21,103],"noise":[23,55,77,127],"distribution":[24,56,78],"crucial":[26],"NCE's":[28,71],"performance.":[29],"However,":[30],"such":[31],"observations":[32],"have":[33],"never":[34],"made":[36],"formal":[37],"or":[38,59],"quantitative.":[39],"In":[40,63],"fact,":[41],"it":[42],"not":[44],"even":[45],"clear":[46],"whether":[47],"difficulties":[49],"arising":[50],"from":[51],"poorly":[53],"chosen":[54],"are":[57,129],"statistical":[58],"algorithmic":[60],"in":[61,130],"nature.":[62],"this":[64],"work,":[65],"we":[66,82,99],"formally":[67],"pinpoint":[68],"reasons":[69],"poor":[72],"performance":[73],"when":[74,123],"an":[75,89,109],"inappropriate":[76],"used.":[80],"Namely,":[81],"prove":[83],"these":[84],"challenges":[85],"arise":[86],"due":[87],"to":[88],"ill-behaved":[90],"(more":[91],"precisely,":[92],"flat)":[93],"loss":[94,111],"landscape.":[95],"To":[96],"address":[97],"this,":[98],"introduce":[100],"variant":[102],"NCE":[104],"called":[105],"\"eNCE\"":[106],"which":[107,114],"uses":[108],"exponential":[110,133],"and":[112,126],"normalized":[115],"gradient":[116],"descent":[117],"addresses":[118],"landscape":[120],"issues":[121],"provably":[122],"target":[125],"distributions":[128],"given":[132],"family.":[134]},"counts_by_year":[],"updated_date":"2026-10-16T09:59:22.235173","created_date":"2021-10-25T00:00:00"}