{"id":"https://openalex.org/W7095147207","doi":null,"title":"Enhanced gradient for learning boltzmann machines (abstract","display_name":"Enhanced gradient for learning boltzmann machines (abstract","publication_year":2011,"publication_date":"2011-01-01","ids":{"openalex":"https://openalex.org/W7095147207"},"language":"en","primary_location":{"id":"pmh:oai:CiteSeerX.psu:10.1.1.387.7318","doi":null,"is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.387.7318","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://snowbird.djvuzone.org/2011/abstracts/109.pdf","raw_type":"text"},"volume":null,"type":"article","indexed_in":[],"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/A5008894339","display_name":"Tapani Raiko","orcid":"https://orcid.org/0000-0002-0321-304X","observed_orcids":["https://orcid.org/0000-0002-0321-304X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tapani Raiko","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091175785","display_name":"Kyunghyun Cho","orcid":"https://orcid.org/0000-0003-1669-3211","observed_orcids":["https://orcid.org/0000-0003-1669-3211"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kyunghyun Cho","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5142987959","display_name":"Er Ilin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Er Ilin","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.5108396,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.9829000234603882,"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/T10320","display_name":"Neural Networks and Applications","score":0.9829000234603882,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.006000000052154064,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.0010000000474974513,"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/boltzmann-machine","display_name":"Boltzmann machine","score":0.8389999866485596},{"id":"https://openalex.org/keywords/restricted-boltzmann-machine","display_name":"restricted Boltzmann machine","score":0.9390000104904175},{"id":"https://openalex.org/keywords/greedy-learning","display_name":"greedy learning","score":0.49300000071525574},{"id":"https://openalex.org/keywords/learning-rate-schedules","display_name":"learning rate scheduling","score":0.746999979019165},{"id":"https://openalex.org/keywords/weight-initialization","display_name":"weight initialization","score":0.7580000162124634},{"id":"https://openalex.org/keywords/training-plateau","display_name":"training plateau","score":0.5479999780654907}],"concepts":[{"id":"https://openalex.org/C192576344","wikidata":"https://www.wikidata.org/wiki/Q194706","display_name":"Boltzmann machine","level":3,"score":0.7347999811172485},{"id":"https://openalex.org/C35304006","wikidata":"https://www.wikidata.org/wiki/Q5962","display_name":"Boltzmann constant","level":2,"score":0.5547000169754028},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4799000024795532},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4693000018596649},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4636000096797943},{"id":"https://openalex.org/C178650346","wikidata":"https://www.wikidata.org/wiki/Q201984","display_name":"Covariance","level":2,"score":0.43380001187324524},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43299999833106995},{"id":"https://openalex.org/C190470478","wikidata":"https://www.wikidata.org/wiki/Q2370229","display_name":"Invariant (physics)","level":2,"score":0.41110000014305115},{"id":"https://openalex.org/C199354608","wikidata":"https://www.wikidata.org/wiki/Q7316287","display_name":"Restricted Boltzmann machine","level":3,"score":0.39820000529289246},{"id":"https://openalex.org/C137800194","wikidata":"https://www.wikidata.org/wiki/Q11713455","display_name":"Interpolation (computer graphics)","level":3,"score":0.3718000054359436},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.3434999883174896},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.3328000009059906},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.298799991607666},{"id":"https://openalex.org/C115903097","wikidata":"https://www.wikidata.org/wiki/Q7094097","display_name":"Online machine learning","level":3,"score":0.28029999136924744},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.274399995803833},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.2635999917984009},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25940001010894775}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:CiteSeerX.psu:10.1.1.387.7318","doi":null,"is_oa":false,"landing_page_url":"http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.387.7318","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://snowbird.djvuzone.org/2011/abstracts/109.pdf","raw_type":"text"}],"best_oa_location":null,"sustainable_development_goals":[],"sustainable_development_goals_aurora":[{"score":0.42228323221206665,"id":"https://metadata.un.org/sdg/8","display_name":"Decent work and economic growth"}],"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/W7100633568","https://openalex.org/W7100347142","https://openalex.org/W2200708944","https://openalex.org/W2132023809","https://openalex.org/W7096961648","https://openalex.org/W6990315052","https://openalex.org/W7096103107","https://openalex.org/W2598354014","https://openalex.org/W1816783926","https://openalex.org/W7095437728"],"abstract_inverted_index":{"Boltzmann":[0,23,61],"machines":[1],"are":[2,52,81],"often":[3,32],"used":[4],"as":[5,21,154],"building":[6],"blocks":[7],"in":[8],"greedy":[9],"learning":[10,30,41,79],"of":[11,39,47,90,128,144],"deep":[12],"networks.":[13],"However,":[14],"training":[15,92],"even":[16,103],"a":[17],"simplified":[18],"model,":[19],"known":[20],"restricted":[22],"machine,":[24],"can":[25,63,96],"be":[26,64],"extremely":[27],"laborious:":[28],"Traditional":[29],"algorithms":[31,95],"converge":[33],"only":[34],"with":[35],"the":[36,40,45,48,72,123,126,130,134,139],"right":[37],"choice":[38],"rate":[42],"scheduling":[43],"and":[44,70,74],"scale":[46],"initial":[49],"weights.":[50],"They":[51],"also":[53],"sensitive":[54],"to":[55,84,118],"specific":[56],"data":[57],"representation:":[58],"An":[59],"equivalent":[60],"machine":[62],"obtained":[65],"by":[66],"flipping":[67],"some":[68],"bits":[69],"changing":[71],"weights":[73],"biases":[75],"accordingly,":[76],"but":[77],"traditional":[78,94],"rules":[80,150],"not":[82],"invariant":[83,117],"such":[85,113],"transformations.":[86,120],"Without":[87],"careful":[88],"tuning":[89],"these":[91],"settings,":[93],"easily":[97],"get":[98],"stuck":[99],"at":[100],"plateaus":[101],"or":[102],"diverge.":[104],"We":[105],"propose":[106],"an":[107],"enhanced":[108,148],"gradient":[109],"which":[110],"is":[111,116,122],"derived":[112],"that":[114],"it":[115],"bit-flipping":[119],"It":[121],"interpolation":[124],"between":[125],"updates":[127],"all":[129],"transformed":[131],"models,":[132],"where":[133],"weight":[135],"update":[136,149],"depends":[137],"on":[138],"covariance":[140],"rather":[141],"than":[142],"products":[143],"activations.":[145],"The":[146],"final":[147],"are,":[151],"then,":[152],"formulated":[153],"\u02dc\u2207wij":[155],"=":[156,166],"covd":[157],"(xi,":[158,162],"xj)":[159,163],"\u2212":[160],"covm":[161],"(1)":[164],"\u02dc\u2207bi":[165],"\u2329xi\u232a":[167],"d":[168]},"counts_by_year":[],"updated_date":"2026-10-14T16:39:01.820549","created_date":"2025-10-27T00:00:00"}