Quantized Kernel Recursive Least Squares Algorithm

In a recent paper, we developed a novel quantized kernel least mean square algorithm, in which the input space is quantized (partitioned into smaller regions) and the network size is upper bounded by the quantization codebook size (number of the regions). In this paper, we propose the quantized kern...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transaction on neural networks and learning systems Ročník 24; číslo 9; s. 1484 - 1491
Hlavní autoři:	Chen, Badong, Zhao, Songlin, Zhu, Pingping, Principe, Jose C.
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York, NY IEEE 01.09.2013 Institute of Electrical and Electronics Engineers The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Algorithms Applied sciences Bioinformatics Coding, codes Computer science; control theory; systems Computer simulation Controllability Data processing. List processing. Character string processing Exact sciences and technology Information, signal and communications theory Kernel Kernel recursive least squares (KRLS) Kernels Learning systems Least squares method Manganese Memory organisation. Data processing On-line systems Partitioning algorithms quantization Quantization (signal) quantized kernel recursive least squares (QKRLS) Recursive Regression Signal and communications theory Software sparsification Telecommunications and information theory Vector quantization Cluster analysis Monte Carlo method Vector quantization Updating Regression analysis Vector method Recursive algorithm Modeling Kernel method Upper bound Codebook Kernel recursive least squares (KRLS) Least squares method Signal quantization Optimal solution sparsification Sparse representation quantized kernel recursive least squares (QKRLS) quantization
ISSN:	2162-237X, 2162-2388, 2162-2388
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	In a recent paper, we developed a novel quantized kernel least mean square algorithm, in which the input space is quantized (partitioned into smaller regions) and the network size is upper bounded by the quantization codebook size (number of the regions). In this paper, we propose the quantized kernel least squares regression, and derive the optimal solution. By incorporating a simple online vector quantization method, we derive a recursive algorithm to update the solution, namely the quantized kernel recursive least squares algorithm. The good performance of the new algorithm is demonstrated by Monte Carlo simulations.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 ObjectType-Article-2 ObjectType-Feature-1 content type line 23
ISSN:	2162-237X 2162-2388 2162-2388
DOI:	10.1109/TNNLS.2013.2258936