Convergence rates for decentralized consistent location parameter estimation in the presence of Gaussian outliers.
Hakan DeliçPublished in: Signal Process. (1997)
Keyphrases
- machine learning
- parameter estimation
- convergence rate
- maximum likelihood
- model selection
- gaussian kernels
- least squares
- learning rate
- markov random field
- expectation maximization
- em algorithm
- primal dual
- parameter estimation algorithm
- parameter values
- maximum likelihood estimation
- random fields
- numerical stability
- gaussian kernel
- approximate inference
- covariance matrices
- parameter estimates
- estimation problems
- low memory requirements
- markov fields
- gaussian mixture model
- bayesian networks
- computer vision
- model fitting
- scale space