Abstract | ||
---|---|---|
This paper focuses on the development of model dimension/order determination algorithms for determining minimal dimensions/orders of recurrent neural networks using only input-output measurements of unknown systems. We present two types of model dimension/order determination approaches. The first type is named all-in-one strategy that includes the minimum description length (MDL) principle and the eigensystem realization algorithm (ERA). This type is capable of identifying the model dimension/order and model parameters simultaneously. The other type is named divide-and-conquer strategy that includes the Lipschitz quotients and false nearest neighbors (FNN). This type usually requires additional parameter optimization algorithms to estimate the model parameters for closely emulating the dynamic behavior of unknown systems. The effectiveness of these four algorithms has been validated through nonlinear dynamic system identification examples. In addition, we provide performance comparisons and discussion on the characteristics of these four algorithms as method-selection guidelines. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1016/j.patrec.2008.05.007 | Pattern Recognition Letters |
Keywords | Field | DocType |
recurrent neural networks,nonlinear dynamic system identification,minimal dimension,divide-and-conquer strategy,dynamic behavior,nonlinear system identification,all-in-one strategy,model parameter,order determination approach,model dimension/order determination,minimal model dimension,minimal realization,order determination algorithm,recurrent neural network,model dimension,unknown system,divide and conquer,nearest neighbor,input output,minimum description length | Eigensystem realization algorithm,Order dimension,Minimum description length,Recurrent neural network,Algorithm,Nonlinear system identification,Lipschitz continuity,System identification,Minimal realization,Mathematics | Journal |
Volume | Issue | ISSN |
30 | 9 | Pattern Recognition Letters |
Citations | PageRank | References |
0 | 0.34 | 5 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jeen-shing Wang | 1 | 729 | 53.58 |
Yu-liang Hsu | 2 | 166 | 16.16 |
Hung-Yi Lin | 3 | 39 | 8.74 |
Yen-ping Chen | 4 | 219 | 15.27 |