Parallelizing Training of Deep Generative Models on Massive Scientific Datasets.
Sam Ade JacobsBrian Van EssenDavid HysomJae-Seung YeomTim MoonRushil AnirudhJayaraman J. ThiagarajanShusen LiuPeer-Timo BremerJim GaffneyTom BensonPeter B. RobinsonJ. Luc PetersonBrian K. SpearsPublished in: CoRR (2019)
Keyphrases
- generative model
- deep belief networks
- restricted boltzmann machine
- probabilistic model
- discriminatively trained
- discriminative learning
- hierarchical hidden markov models
- mixture model
- maximum entropy principle
- discriminative models
- em algorithm
- conditional random fields
- deep learning
- semi supervised
- prior knowledge
- expectation maximization
- data sets
- topic models
- pairwise
- training set
- training samples
- naive bayes models
- representational power
- hierarchical models
- object detection
- bayesian networks
- clustering algorithm
- information retrieval