A memory optimal BFGS neural network training algorithm

Proceedings of the International Joint Conference on Neural Networks

Volumn 1, Issue , 2002, Pages 513-518

  McLoone, Seán F ^a   Asirvadam, Vijanth S ^a   Irwin, George W ^a  

^a QUEEN'S UNIVERSITY BELFAST   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

ITERATIVE METHODS; MATHEMATICAL MODELS; MATRIX ALGEBRA; OPTIMIZATION; VECTORS;

MEMORY OPTIMAL NEURAL NETWORK TRAINING ALGORITHM; VARIABLE MEMORY;

FEEDFORWARD NEURAL NETWORKS;

EID: 0036073581     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (17)

1. Learning internal representations by error propagation
2. A scaled conjugate algorithm for fast supervised learning
3. First and second-order methods for learning
4. Minimisation methods for training feedforward neural networks
5. Training feedforward networks with Marquardt algorithm
6. Fast gradient based offline training of multilayer perceptrons
7. A variable memory quasi-newton training algorithm
8. QN-like variable storage conjugate gradients
9. A relationship between the BFGS and conjugate gradient algorithm and its implications for new algorithms
10. Updating Quasi-Newton matrices with limited storage
11. An efficient learning rate selection strategy for gradient based neural network training algorithms
12. Fast parallel off-line training of multilayer perceptrons
13. A memory optimal neural network training algorithm
14. A parallel variable memory BFGS training algorithm
15. On the limited memory BFGS method for large scale optimization
16. Robust approximately time-optimal fuzzy control of a two-tank system
17. Neural network training for modelling and control