Learning Deep Generative Models

By Ruslan Salakhutdinov. 2009

454 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

A Kernelized Stein Discrepancy for Goodness-of-fit Tests (2016, background): “Unfortunately, modern learning techniques increasingly involve complex probabilistic models with computationally intractable likelihoods or CDFs, such as large graphical models, hidden variables models and deep generative models (Koller & Friedman, 2009; Salakhutdinov, 2015).”
Parallel tempering is efficient for learning restricted Boltzmann machines (2010, methodology): “For estimating the normalizing constant, this paper adapts the method utilizingannealed importance sampling (AIS) [10] which has been successfully adapted for computing the normalizing constant of RBM [11].”
DNN-based stochastic postfilter for HMM-based speech synthesis (2014, background): “, deep belief networks (DBNs) [11, 12] or deep Boltzmann machines (DBMs) [13], to construct deeper mapping relationships.”
A Deep Generative Architecture for Postfiltering in Statistical Parametric Speech Synthesis (2015, methodology): “Therefore, the contrastive divergence (CD) algorithm is usually used to estimate the parameters of an RBM [29], [30] and the annealed importance sampling (AIS) algorithm is adopted to approximate the partition function for model evaluation [31].”
A systematic literature review on machine learning applications for sustainable agriculture supply chain performance (2020, unknown): “Deep ANNs are referred to as Deep Learning because of multiple hidden layers (Salakhutdinov, 2015; Araque et al. 2017).”
Enhanced Gradient and Adaptive Learning Rate for Training Restricted Boltzmann Machines (2011, methodology): “Annealed importance sampling (AIS) provides a way to estimate the normalizing constant of RBM (Salakhutdinov, 2009a).”
Autoencoder-based feature learning for cyber security applications (2017, background): “The motivation is that AE as a generative model is capable of learning a reasonable notion of semantic similarity and the relation among input features [10, 11] .”
Modeling spectral envelopes using restricted Boltzmann machines for statistical parametric speech synthesis (2013, methodology): “In order to make minimum modification to the original model training and parameter generation procedures, the RBM-based spectral envelope modeling method is implemented as a post-processing step performed on the trained context-dependent Gaussian HMMs using the conventional spectral parameters such…”
Artificial intelligence in drug discovery: recent advances and future perspectives (2021, background): “Contemporary approaches are often based on generative deep learning models [124], which sample new molecules from a learned latent molecular representation.”