An Exponentially Increasing Step-size for Parameter Estimation in Statistical Models.
Nhat HoTongzheng RenSujay SanghaviPurnamrita SarkarRachel WardPublished in: CoRR (2022)
Keyphrases
- parameter estimation
- statistical models
- step size
- cost function
- convergence rate
- statistical modeling
- maximum likelihood
- convergence speed
- least squares
- model selection
- markov random field
- em algorithm
- approximate inference
- maximum likelihood estimation
- random fields
- expectation maximization
- parameter estimation algorithm
- variable step size
- markov chain monte carlo
- multiscale
- exponential family
- image segmentation
- computer vision
- differential evolution
- higher order
- feature vectors
- objective function