Improved pre-training of Deep Belief Networks using Sparse Encoding Symmetric Machines.
Christian PlahlTara N. SainathBhuvana RamabhadranDavid NahamooPublished in: ICASSP (2012)
Keyphrases
- deep belief networks
- restricted boltzmann machine
- sparse coding
- deep architectures
- unsupervised learning
- hierarchical models
- deep learning
- generative model
- linear combination
- neural network
- sparse representation
- natural images
- probabilistic model
- high dimensional
- training set
- learning algorithm
- hierarchical model
- structured prediction
- reinforcement learning
- machine learning