The application of an oblique-projected Landweber method to a model of supervised learning.
Björn JohanssonTommy ElfvingVladimir KozlovYair CensorPer-Erik ForssénGösta H. GranlundPublished in: Math. Comput. Model. (2006)
Keyphrases
- mathematical model
- statistical model
- modeling method
- probabilistic model
- optimization method
- classification method
- objective function
- hybrid model
- network model
- linear model
- prior information
- prediction model
- theoretical analysis
- gaussian distribution
- bp neural network
- evaluation method
- detection method
- cost function
- study proposes
- prior knowledge
- similarity measure
- test data
- sensitivity analysis
- monte carlo simulation
- high order
- closed form
- classification algorithm
- training data
- unsupervised learning
- supervised learning
- statistical methods
- decision trees
- recognition process
- reconstruction method
- reinforcement learning
- support vector machine
- recognition algorithm
- parameter estimation
- parameter space
- input data
- model selection
- significant improvement
- regression analysis
- evaluation model
- learning tasks
- neural network