Matches in SemOpenAlex for { <https://semopenalex.org/work/W122577237> ?p ?o ?g. }
- W122577237 endingPage "219" @default.
- W122577237 startingPage "201" @default.
- W122577237 abstract "A deep Boltzmann machine (DBM) is a recently introduced Markov random field model that has multiple layers of hidden units. It has been shown empirically that it is difficult to train a DBM with approximate maximum-likelihood learning using the stochastic gradient unlike its simpler special case, restricted Boltzmann machine (RBM). In this paper, we propose a novel pretraining algorithm that consists of two stages; obtaining approximate posterior distributions over hidden units from a simpler model and maximizing the variational lower-bound given the fixed hidden posterior distributions. We show empirically that the proposed method overcomes the difficulty in training DBMs from randomly initialized parameters and results in a better, or comparable, generative model when compared to the conventional pretraining algorithm." @default.
- W122577237 created "2016-06-24" @default.
- W122577237 creator A5008894339 @default.
- W122577237 creator A5030087918 @default.
- W122577237 creator A5061369039 @default.
- W122577237 creator A5085358291 @default.
- W122577237 date "2015-01-01" @default.
- W122577237 modified "2023-09-26" @default.
- W122577237 title "How to Pretrain Deep Boltzmann Machines in Two Stages" @default.
- W122577237 cites W115070352 @default.
- W122577237 cites W1498436455 @default.
- W122577237 cites W1598712055 @default.
- W122577237 cites W2013035813 @default.
- W122577237 cites W2051144468 @default.
- W122577237 cites W2054912984 @default.
- W122577237 cites W2071709160 @default.
- W122577237 cites W2096192494 @default.
- W122577237 cites W2100495367 @default.
- W122577237 cites W2103819961 @default.
- W122577237 cites W2110798204 @default.
- W122577237 cites W2112796928 @default.
- W122577237 cites W2116064496 @default.
- W122577237 cites W2116825644 @default.
- W122577237 cites W2132023809 @default.
- W122577237 cites W2136922672 @default.
- W122577237 cites W2136936677 @default.
- W122577237 cites W2150529939 @default.
- W122577237 cites W2163922914 @default.
- W122577237 cites W2171312459 @default.
- W122577237 cites W3106144113 @default.
- W122577237 cites W4254253786 @default.
- W122577237 doi "https://doi.org/10.1007/978-3-319-09903-3_10" @default.
- W122577237 hasPublicationYear "2015" @default.
- W122577237 type Work @default.
- W122577237 sameAs 122577237 @default.
- W122577237 citedByCount "2" @default.
- W122577237 countsByYear W1225772372014 @default.
- W122577237 countsByYear W1225772372015 @default.
- W122577237 crossrefType "book-chapter" @default.
- W122577237 hasAuthorship W122577237A5008894339 @default.
- W122577237 hasAuthorship W122577237A5030087918 @default.
- W122577237 hasAuthorship W122577237A5061369039 @default.
- W122577237 hasAuthorship W122577237A5085358291 @default.
- W122577237 hasBestOaLocation W1225772372 @default.
- W122577237 hasConcept C105795698 @default.
- W122577237 hasConcept C108583219 @default.
- W122577237 hasConcept C11413529 @default.
- W122577237 hasConcept C115961682 @default.
- W122577237 hasConcept C121332964 @default.
- W122577237 hasConcept C124504099 @default.
- W122577237 hasConcept C153180895 @default.
- W122577237 hasConcept C154945302 @default.
- W122577237 hasConcept C167966045 @default.
- W122577237 hasConcept C192576344 @default.
- W122577237 hasConcept C199354608 @default.
- W122577237 hasConcept C202444582 @default.
- W122577237 hasConcept C23224414 @default.
- W122577237 hasConcept C2778045648 @default.
- W122577237 hasConcept C33923547 @default.
- W122577237 hasConcept C35304006 @default.
- W122577237 hasConcept C39890363 @default.
- W122577237 hasConcept C41008148 @default.
- W122577237 hasConcept C49781872 @default.
- W122577237 hasConcept C9652623 @default.
- W122577237 hasConcept C97355855 @default.
- W122577237 hasConceptScore W122577237C105795698 @default.
- W122577237 hasConceptScore W122577237C108583219 @default.
- W122577237 hasConceptScore W122577237C11413529 @default.
- W122577237 hasConceptScore W122577237C115961682 @default.
- W122577237 hasConceptScore W122577237C121332964 @default.
- W122577237 hasConceptScore W122577237C124504099 @default.
- W122577237 hasConceptScore W122577237C153180895 @default.
- W122577237 hasConceptScore W122577237C154945302 @default.
- W122577237 hasConceptScore W122577237C167966045 @default.
- W122577237 hasConceptScore W122577237C192576344 @default.
- W122577237 hasConceptScore W122577237C199354608 @default.
- W122577237 hasConceptScore W122577237C202444582 @default.
- W122577237 hasConceptScore W122577237C23224414 @default.
- W122577237 hasConceptScore W122577237C2778045648 @default.
- W122577237 hasConceptScore W122577237C33923547 @default.
- W122577237 hasConceptScore W122577237C35304006 @default.
- W122577237 hasConceptScore W122577237C39890363 @default.
- W122577237 hasConceptScore W122577237C41008148 @default.
- W122577237 hasConceptScore W122577237C49781872 @default.
- W122577237 hasConceptScore W122577237C9652623 @default.
- W122577237 hasConceptScore W122577237C97355855 @default.
- W122577237 hasLocation W1225772371 @default.
- W122577237 hasLocation W1225772372 @default.
- W122577237 hasOpenAccess W122577237 @default.
- W122577237 hasPrimaryLocation W1225772371 @default.
- W122577237 hasRelatedWork W122577237 @default.
- W122577237 hasRelatedWork W1835073532 @default.
- W122577237 hasRelatedWork W1974618110 @default.
- W122577237 hasRelatedWork W2164700406 @default.
- W122577237 hasRelatedWork W2171312459 @default.
- W122577237 hasRelatedWork W2503569529 @default.
- W122577237 hasRelatedWork W2588503761 @default.
- W122577237 hasRelatedWork W2598231273 @default.