On Objective Function, Regularizer, and Prediction Error of a Learning Algorithm for Dealing With Multiplicative Weight Noise.
J. P. F. SumChi-Sing LeungKevin I.-J. HoPublished in: IEEE Trans. Neural Networks (2009)
Keyphrases
- prediction error
- objective function
- learning algorithm
- motion compensated
- bit rate
- inter frame
- multi task learning
- regularization term
- active learning
- semi supervised
- power spectral density
- linear prediction
- constrained optimization
- machine learning
- signal to noise ratio
- linear predictors
- motion vectors
- noise reduction
- back propagation
- optimization problems
- weight vector
- optimal solution
- image processing
- reversible watermarking
- learning problems
- three dimensional