Convex and concave envelopes of artificial neural network activation functions for deterministic global optimization.
Matthew E. WilhelmChenyu WangMatthew D. StuberPublished in: J. Glob. Optim. (2023)
Keyphrases
- global optimization
- activation function
- artificial neural networks
- piecewise linear
- back propagation
- neural network
- continuous functions
- feed forward
- hidden layer
- global optimality
- convex functions
- particle swarm optimization
- multilayer perceptron
- radial basis function
- objective function
- pso algorithm
- convex optimization
- neural nets
- neural architecture
- genetic algorithm
- learning rate
- multi layer perceptron
- globally optimal
- network architecture
- rbf neural network
- dynamic programming
- multi objective
- clustering algorithm
- feature selection
- deterministic annealing
- imperialist competitive algorithm