Early Stage Convergence and Global Convergence of Training Mildly Parameterized Neural Networks.
Mingze WangChao MaPublished in: CoRR (2022)
Keyphrases
- global convergence
- early stage
- neural network
- convergence rate
- convergence speed
- convergence analysis
- global optimum
- training process
- optimization methods
- training algorithm
- constrained optimization problems
- convex minimization
- line search
- globally convergent
- feedforward neural networks
- multi layer perceptron
- coordinate ascent
- back propagation
- backpropagation algorithm
- particle swarm
- fuzzy logic
- hybrid algorithm
- neural network training
- artificial neural networks
- multi layer
- objective function
- faster convergence
- optimization method
- genetic algorithm