Stochastic Natural Gradient Descent by estimation of empirical covariances.
Luigi MalagòMatteo MatteucciGiovanni PistonePublished in: IEEE Congress on Evolutionary Computation (2011)
Keyphrases
- estimation algorithm
- cost function
- stochastic optimization
- theoretical analysis
- loss function
- covariance matrix
- real time
- real world
- multiscale
- stochastic model
- computationally efficient
- parameter estimation
- maximum likelihood estimation
- maximum a posteriori estimation
- estimation accuracy
- monte carlo
- least squares
- objective function
- learning algorithm
- neural network