Batch-mode identification of black-box models using feedforward neural networks

Proceedings of the American Control Conference

Volumn 1, Issue , 2002, Pages 406-411

  Alessandri, A ^a   Sanguineti, M ^b   Maggiore, M ^c  

^a CNR   (Italy)

^b UNIVERSITY OF GENOA   (Italy)

^c UNIVERSITY OF TORONTO   (Canada)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER SIMULATION; DISCRETE TIME CONTROL SYSTEMS; FEEDFORWARD NEURAL NETWORKS; LEARNING ALGORITHMS; NONLINEAR SYSTEMS; OPTIMIZATION; PARAMETER ESTIMATION;

BATCH-MODE IDENTIFICATION; DISCRETE-TIME NONLINEAR SYSTEM;

IDENTIFICATION (CONTROL SYSTEMS);

EID: 0036057179     PISSN: 07431619     EISSN: None     Source Type: Journal    
DOI: 10.1109/ACC.2002.1024839     Document Type: Article

References (11)

1. Nonlinear black-box models in system identification: Mathematical foundations
2. Approximation and learning unknown mappings using multilayer feedforward networks with bounded weights
3. Some new results on neural network approximation
4. Approximation of functions by perceptron networks with bounded number of hidden units
5. Neural networks as nonlinear approximators
6. Practical identification of NAR-MAX models using radial basis functions
7. Input-output modeling of nonlinear systems with time-varying models
8. A neural state estimator with bounded errors for nonlinear systems
9. Optimization-based learning with bounded error for feedforward neural networks
10. A theoretical comparison of batch-mode, on-line, cyclic, and almost-cyclic learning
11. On overfitting, generalization, and randomly expanded training sets