Academic
Publications
Incremental learning of dynamic fuzzy neural networks for accurate system modeling

Incremental learning of dynamic fuzzy neural networks for accurate system modeling,10.1016/j.fss.2008.09.005,Fuzzy Sets and Systems,Xingsheng Deng,Xin

Incremental learning of dynamic fuzzy neural networks for accurate system modeling   (Citations: 11)
BibTex | RIS | RefWorks Download
In this paper we propose a novel incremental learning approach based on a hybrid fuzzy neural net framework. A key feature of the approach is the adaptation of the fuzzy neural network (FNN) modeling to every new data. The typical algorithm of FNN is inefficient when used in an accurate online time series because they must be retrained from scratch every time the training set is modified. In order to reduce the expense of FNN learning for a dynamic system, a general methodology leading to quick algorithms for FNN modeling is developed. The FNN-LM algorithm for a static FNN and incremental learning algorithm (ILA) for dynamic fuzzy neural network (DFNN) are also presented to enforce the model to approximate every new sample. The ILA approach has the advantages of avoiding increasing the ranks of matrixes and avoiding solving the inverse matrix when samples increase gradually. When it is used to predict an accurate online time series, the DFNN model can efficiently update a trained static FNN with a very fast speed according to the sample added to the training set. Numerical experiments validate our theoretical results. Excellent performances of the proposed approach in modeling accuracy and learning convergence are exhibited.
Journal: Fuzzy Sets and Systems - FSS , vol. 160, no. 7, pp. 972-987, 2009
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...To evaluate the performance of the FRI-FCM, we tested it on five datasets and compared the results with other recently developed fuzzysystemsandFNNs[32],[33],[39]‐[41],[43]‐[ 46], [49], [51]‐[53]...
    • ...This problem has been used as a benchmark problem in the areas of NNs, fuzzy systems, and hybrid systems [31], [51]‐[53]...
    • ... developed fuzzy systems and FNNs, including the hybridneuralfuzzyinferencesystem(HyFIS)[46],theneurofuzzy function approximator (NEFPROX) [47], the dynamic FNN (D-FNN) [48], a combination of the genetic algorithm and gradient descent algorithm (GEFREX) [49], the generalized FNN (G-FNN) [50], the type-2 fuzzy logic system (T2FLS) [51], the self-evolving interval type-2 FNN (SEIT2FNN) [52], the incremental learning algorithm (ILA) [53], ...

    Hengjie J. Songet al. An Extension to Fuzzy Cognitive Maps for Classification and Prediction

    • ...Because accurate forecasting for the future trend is usually difficult in complex and nonlinear real-world problems, many researchers have used intelligent computing methods for time series forecasting, where fuzzy theory and neural networks have been widely investigated [1]-[4]...

    Chunshien Liet al. Complex Fuzzy Computing to Time Series Prediction A Multi-Swarm PSO Le...

    • ...The structure of the fuzzy neural networks with five layers is given in [2], [3], which are also shown in Fig. 2...

    Yong-song Luet al. An improved Fuzzy neural networks algorithm and its application in res...

Sort by: