On-line sliding-window Levenberg-Marquardt methods for neural network models
On-line learning algorithms are needed when the process to be modeled is time-varying or when it is impossible to obtain off-line data that covers the whole operating region. To minimize the problems of parameter shadowing and interference, sliding-based algorithms are used. It is shown that, by usi...
| Autor principal: | |
|---|---|
| Outros Autores: | |
| Formato: | conferenceObject |
| Idioma: | eng |
| Publicado em: |
2013
|
| Texto completo: | http://hdl.handle.net/10400.1/2252 |
| País: | Portugal |
| Oai: | oai:sapientia.ualg.pt:10400.1/2252 |
| Resumo: | On-line learning algorithms are needed when the process to be modeled is time-varying or when it is impossible to obtain off-line data that covers the whole operating region. To minimize the problems of parameter shadowing and interference, sliding-based algorithms are used. It is shown that, by using a sliding window policy that enforces the novelty of data stored in the sliding window, and by using a procedure to prevent unnecessary parameter updates, the performance achieved is improved over a FIFO policy with fixed parameter updates. Important savings in computational effort are also obtained. |
|---|