.png){kind=icon}
Return
Quantized kernel Lleast lncosh algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.sigpro.2021.108255.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper introduces the kernel least lncosh (KLL) algorithm, in which the lncosh (logarithm of hyperbolic cosine) cost function is successfully applied in the reproducing-kernel-Hilbert space. The online vector quantization (VQ) is then used to quantize the input space to construct the algorithm which can curb the growth of network size. As a result, the quantized kernel least lncosh (QKLL) algorithm is developed, which is robust in non-Gaussian environments. The sufficient condition for mean-square convergence of the QKLL algorithm has been conducted. The performance of the KLL and QKLL algorithms is demonstrated by the short-term chaotic time-series prediction and non-linear channel-equalization (NCE). (c) 2021 Elsevier B.V. All rights reserved.
Keywords:
Kernel least lncosh
Reproducing-kernel-Hilbert space
Vector quantization
Quantized kernel least lncosh
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.6
Papers:
9.9K
Citations:
1.7W
Organization
Cited Papers
.png){kind=icon}
ShareSaveShareSaveShareSaveShareSaveShareSave
ShareSave
View more