Fast Parallel Expectation Maximization for Gaussian Mixture Models on GPUs Using CUDA.
N. S. L. Phani KumarSanjiv SatoorIan BuckPublished in: HPCC (2009)
Keyphrases
- gaussian mixture model
- expectation maximization
- graphics processing units
- em algorithm
- mixture model
- parallel programming
- parallel processing
- general purpose
- maximum likelihood
- probabilistic model
- gaussian mixture
- speaker recognition
- unsupervised learning
- parallel computing
- parallel architectures
- bayesian framework
- density estimation
- gaussian distribution
- maximum likelihood estimation
- covariance matrices
- speaker identification
- finite mixtures
- image segmentation
- probability density function
- generative model
- shared memory
- probability density
- mixture of gaussians
- parallel implementation
- feature vectors
- variational bayes
- minimum message length
- feature space
- computing systems
- language model
- k means
- finite mixture model