SMEM Algorithm Is Not Fully Compatible with Maximum-Likelihood Framework.
Akihiro MinagawaNorio TagawaToshiyuki TanakaPublished in: Neural Comput. (2002)
Keyphrases
- maximum likelihood
- computational complexity
- estimation algorithm
- objective function
- expectation maximization
- learning algorithm
- cost function
- dynamic programming
- high accuracy
- classification algorithm
- times faster
- preprocessing
- parameter estimation
- experimental evaluation
- segmentation algorithm
- probabilistic model
- matching algorithm
- recognition algorithm
- theoretical analysis
- input data
- improved algorithm
- computational cost
- k means
- search space
- similarity measure
- bayesian framework
- genetic algorithm
- particle swarm optimization
- image reconstruction
- ant colony optimization
- detection algorithm
- optimization algorithm
- worst case
- optimal solution