Necessary and sufficient condition for convergence of iterative learning algorithm.
Sunan HuangKok Kiong TanTong Heng LeePublished in: Autom. (2002)
Keyphrases
- sufficient conditions
- learning algorithm
- linear complementarity problem
- exponential stability
- convergence rate
- asymptotic stability
- learning rate
- back propagation
- asymptotic optimality
- convergence speed
- linear systems
- training data
- machine learning algorithms
- fixed point
- active learning
- lyapunov function
- efficiently computable
- learning tasks
- generalization ability
- machine learning
- equilibrium point
- learning problems
- training samples
- support vector machine
- objective function