{"id":"https://openalex.org/W3198528488","doi":"https://doi.org/10.1109/isit45174.2021.9518141","title":"A Provably Convergent Information Bottleneck Solution via ADMM","display_name":"A Provably Convergent Information Bottleneck Solution via ADMM","publication_year":2021,"publication_date":"2021-07-12","ids":{"openalex":"https://openalex.org/W3198528488","doi":"https://doi.org/10.1109/isit45174.2021.9518141","mag":"3198528488"},"language":"en","primary_location":{"id":"doi:10.1109/isit45174.2021.9518141","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit45174.2021.9518141","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5018629338","display_name":"Teng-Hui Huang","orcid":"https://orcid.org/0000-0001-8200-1178"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Teng-Hui Huang","raw_affiliation_strings":["Purdue University,Department of Electrical and Computer Engineering,West Lafayette,IN,USA","Department of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"],"affiliations":[{"raw_affiliation_string":"Purdue University,Department of Electrical and Computer Engineering,West Lafayette,IN,USA","institution_ids":["https://openalex.org/I219193219"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051018107","display_name":"Aly El Gamal","orcid":"https://orcid.org/0000-0002-0400-4506"},"institutions":[{"id":"https://openalex.org/I219193219","display_name":"Purdue University West Lafayette","ror":"https://ror.org/02dqehb95","country_code":"US","type":"education","lineage":["https://openalex.org/I219193219"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Aly el Gamal","raw_affiliation_strings":["Purdue University,Department of Electrical and Computer Engineering,West Lafayette,IN,USA","Department of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"],"affiliations":[{"raw_affiliation_string":"Purdue University,Department of Electrical and Computer Engineering,West Lafayette,IN,USA","institution_ids":["https://openalex.org/I219193219"]},{"raw_affiliation_string":"Department of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA","institution_ids":["https://openalex.org/I219193219"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5018629338"],"corresponding_institution_ids":["https://openalex.org/I219193219"],"apc_list":null,"apc_paid":null,"fwci":0.9789,"has_fulltext":false,"cited_by_count":7,"citation_normalized_percentile":{"value":0.80778687,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"43","last_page":"48"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.9995999932289124,"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":0.9995999932289124,"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":0.9991999864578247,"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/T12676","display_name":"Machine Learning and ELM","score":0.9990000128746033,"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/initialization","display_name":"Initialization","score":0.5910987257957458},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5299970507621765},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5263893604278564},{"id":"https://openalex.org/keywords/convergence","display_name":"Convergence (economics)","score":0.4915412366390228},{"id":"https://openalex.org/keywords/information-bottleneck-method","display_name":"Information bottleneck method","score":0.4727120101451874},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.42477312684059143},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.42323243618011475},{"id":"https://openalex.org/keywords/combinatorics","display_name":"Combinatorics","score":0.37330228090286255},{"id":"https://openalex.org/keywords/discrete-mathematics","display_name":"Discrete mathematics","score":0.34662723541259766},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.3166329264640808},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.10235857963562012}],"concepts":[{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.5910987257957458},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5299970507621765},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5263893604278564},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.4915412366390228},{"id":"https://openalex.org/C60008888","wikidata":"https://www.wikidata.org/wiki/Q6031013","display_name":"Information bottleneck method","level":3,"score":0.4727120101451874},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.42477312684059143},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.42323243618011475},{"id":"https://openalex.org/C114614502","wikidata":"https://www.wikidata.org/wiki/Q76592","display_name":"Combinatorics","level":1,"score":0.37330228090286255},{"id":"https://openalex.org/C118615104","wikidata":"https://www.wikidata.org/wiki/Q121416","display_name":"Discrete mathematics","level":1,"score":0.34662723541259766},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3166329264640808},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.10235857963562012},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.0},{"id":"https://openalex.org/C50522688","wikidata":"https://www.wikidata.org/wiki/Q189833","display_name":"Economic growth","level":1,"score":0.0},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.0},{"id":"https://openalex.org/C162324750","wikidata":"https://www.wikidata.org/wiki/Q8134","display_name":"Economics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/isit45174.2021.9518141","is_oa":false,"landing_page_url":"https://doi.org/10.1109/isit45174.2021.9518141","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2021 IEEE International Symposium on Information Theory (ISIT)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W135467536","https://openalex.org/W1686946872","https://openalex.org/W1923817890","https://openalex.org/W2022541060","https://openalex.org/W2032199857","https://openalex.org/W2053801139","https://openalex.org/W2103723258","https://openalex.org/W2148986322","https://openalex.org/W2150695437","https://openalex.org/W2163232332","https://openalex.org/W2163651591","https://openalex.org/W2164278908","https://openalex.org/W2171991173","https://openalex.org/W2551153860","https://openalex.org/W2552808051","https://openalex.org/W2739679565","https://openalex.org/W2902719825","https://openalex.org/W2946936191","https://openalex.org/W2953057288","https://openalex.org/W2960183474","https://openalex.org/W2963412800","https://openalex.org/W2964160479","https://openalex.org/W2989653627","https://openalex.org/W2994090936","https://openalex.org/W2995518662","https://openalex.org/W3103727670","https://openalex.org/W3105393233","https://openalex.org/W3128288498","https://openalex.org/W3198528488","https://openalex.org/W4244633107","https://openalex.org/W4293469690","https://openalex.org/W4301909677","https://openalex.org/W6637108112","https://openalex.org/W6640026981","https://openalex.org/W6651608069","https://openalex.org/W6675471587","https://openalex.org/W6682270173","https://openalex.org/W6729906282","https://openalex.org/W6771723691","https://openalex.org/W6790013338"],"related_works":["https://openalex.org/W2622284819","https://openalex.org/W1504394672","https://openalex.org/W4386721405","https://openalex.org/W2381356463","https://openalex.org/W2996506326","https://openalex.org/W2950826591","https://openalex.org/W4300774107","https://openalex.org/W3193949130","https://openalex.org/W2782011872","https://openalex.org/W4298151174"],"abstract_inverted_index":{"The":[0,48],"Information":[1],"bottleneck":[2],"(IB)":[3],"method":[4,206,235],"enables":[5],"optimizing":[6],"over":[7],"the":[8,41,68,75,90,112,120,140,159,172,183,196,203,217,225,248,260],"trade-off":[9],"between":[10],"compression":[11],"of":[12,17,77,142,207,224,227,259],"data":[13],"and":[14,20,23,30,58,133,139,252],"prediction":[15],"accuracy":[16],"learned":[18],"representations,":[19],"has":[21,37,115],"successfully":[22],"robustly":[24],"been":[25],"applied":[26],"to":[27,46,88,158,171,244],"both":[28,129],"supervised":[29],"unsupervised":[31],"representation":[32],"learning":[33],"problems.":[34],"However,":[35],"IB":[36,42,49,113,153,160,173,184,197],"several":[38],"limitations.":[39],"First,":[40],"problem":[43],"is":[44,56,80,175,220,253],"hard":[45],"optimize.":[47],"Lagrangian":[50,114,174,185,188],"<tex":[51,96,100,122,130,134,228],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[52,97,101,123,131,135,229],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\mathcal{L}_{IB}:":[53],"=I(X;Z)-\\beta":[54],"I(Y;Z)$</tex>":[55],"non-convex":[57],"existing":[59,148],"solutions":[60,70,105],"guarantee":[61],"only":[62],"local":[63],"convergence.":[64],"As":[65],"a":[66,78,82,256],"result,":[67],"obtained":[69,106,246],"depend":[71],"on":[72],"initialization.":[73],"Second,":[74],"evaluation":[76],"solution":[79],"also":[81],"challenging":[83],"task.":[84],"Conventionally,":[85],"it":[86],"resorts":[87],"characterizing":[89],"information":[91,238],"plane,":[92],"that":[93,195,216,241],"is,":[94],"plotting":[95],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$I(Y;Z)$</tex>":[98,136],"versus":[99],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$I(X;Z)$</tex>":[102,132],"for":[103,147,255],"all":[104],"from":[107,211],"different":[108],"initial":[109],"points.":[110],"Furthermore,":[111],"phase":[116,127],"transitions":[117],"while":[118],"varying":[119],"multiplier":[121],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">$\\beta$</tex>":[124,230],".":[125,231],"At":[126],"transitions,":[128],"increase":[137],"abruptly":[138],"rate":[141],"convergence":[143],"becomes":[144],"significantly":[145],"slow":[146],"solutions.":[149],"Recent":[150],"works":[151],"with":[152,202],"adopt":[154],"variational":[155],"surrogate":[156],"bounds":[157],"Lagrangian.":[161],"Although":[162],"allowing":[163],"efficient":[164],"optimization,":[165],"how":[166],"close":[167],"are":[168,242],"these":[169],"surrogates":[170],"not":[176],"clear.":[177],"In":[178],"this":[179],"work,":[180],"we":[181,193,214],"solve":[182],"using":[186],"augmented":[187,191],"methods.":[189],"With":[190],"variables,":[192],"show":[194],"objective":[198],"can":[199],"be":[200],"solved":[201],"alternating":[204],"direction":[205],"multipliers":[208],"(ADMM).":[209],"Different":[210],"prior":[212],"works,":[213],"prove":[215],"proposed":[218],"algorithm":[219],"consistently":[221],"convergent,":[222],"regardless":[223],"value":[226],"Empirically,":[232],"our":[233],"gradient-descent-based":[234],"results":[236],"in":[237],"plane":[239],"points":[240],"comparable":[243],"those":[245],"through":[247],"conventional":[249],"Blahut-Arimoto-based":[250],"solvers,":[251],"convergent":[254],"wider":[257],"range":[258],"penalty":[261],"coefficient":[262],"than":[263],"previous":[264],"ADMM-based":[265],"solvers.":[266]},"counts_by_year":[{"year":2024,"cited_by_count":2},{"year":2023,"cited_by_count":1},{"year":2022,"cited_by_count":3},{"year":2021,"cited_by_count":1}],"updated_date":"2026-03-14T06:41:57.775601","created_date":"2025-10-10T00:00:00"}