翻訳と辞書 Words near each other ・ Kernel preemption ・ Kernel principal component analysis ・ Kernel random forest ・ Kernel regression ・ Kernel relocation ・ Kernel same-page merging ・ Kernel Scheduled Entities ・ Kernel smoother ・ Kernel Transaction Manager ・ Kernel virtual address space ・ Kernel-based Virtual Machine ・ Kernel-Mode Driver Framework ・ Kernel.org ・ KernelCAD ・ KernelCare ・ KernelICA ・ Kernelization ・ Kernell ・ Kernell, California ・ KernelTrap ・ Kernen ・ Kernen (hill) ・ Kernen Omloop Echt-Susteren ・ Kernenried ・ Kerner ・ Kerner (grape) ・ Kerner Commission ・ Kerner House ・ Kerner Optical ・ Kernersville Depot Dictionary Lists mini英和辞書 mini和英辞書 Webster 1913 Latin-English FOLDOC Wikipedia English ウィキペディア

翻訳と辞書　辞書検索 [ 開発暫定版 ] スポンサード リンク KernelICA ： ウィキペディア英語版 KernelICA Kernel independent component analysis (Kernel ICA) is an efficient algorithm for independent component analysis which estimates source components by optimizing a ''generalized variance'' contrast function, which is based on representations in a reproducing kernel Hilbert space. Those contrast functions use the notion of mutual information as a measure of statistical independence. ==Main idea== Kernel ICA is based on the idea that correlations between two random variables can be represented in a reproducing kernel Hilbert space (RKHS), denoted by $\backslash mathcal$, associated with a feature map $L\_x:\; \backslash mathcal\; \backslash mapsto\; \backslash mathbb$ defined for a fixed $x\; \backslash in\; \backslash mathbb$. The $\backslash mathcal$-correlation between two random variables $X$ and $Y$ is defined $\backslash rho\_\}\; \backslash text(\; \backslash langle\backslash \; L\_X,f\; \backslash rangle,\; \backslash langle\backslash \; L\_Y,g\; \backslash rangle)$ where the functions $f,g:\; \backslash mathbb\; \backslash to\; \backslash mathbb$ range over $\backslash mathcal$ and $\backslash text(\; \backslash langle\backslash \; L\_X,f\; \backslash rangle,\; \backslash langle\backslash \; L\_Y,g\; \backslash rangle)\; :=\; \backslash frac\; \backslash text(g(Y))^\; \}$ for fixed $f,g\; \backslash in\; \backslash mathcal$.〔 Note that the reproducing property implies that $f(x)\; =\; \backslash langle\backslash \; L\_x,\; f\; \backslash rangle$ for fixed $x\; \backslash in\; \backslash mathbb$ and $f\; \backslash in\; \backslash mathcal$. It follows then that the $\backslash mathcal$-correlation between two independent random variables is zero. This notion of $\backslash mathcal$-correlations is used for defining ''contrast'' functions that are optimized in the Kernel ICA algorithm. Specifically, if $\backslash mathbf\; :=\; (x\_)\; \backslash in\; \backslash mathbb^$ is a prewhitened data matrix, that is, the sample mean of each column is zero and the sample covariance of the rows is the $m\; \backslash times\; m$ dimensional identity matrix, Kernel ICA estimates a $m\; \backslash times\; m$ dimensional orthogonal matrix $\backslash mathbf$ so as to minimize finite-sample $\backslash mathcal$-correlations between the columns of $\backslash mathbf\; :=\; \backslash mathbf\; \backslash mathbf^$. 抄文引用元・出典: フリー百科事典『 ウィキペディア（Wikipedia）』 ■ウィキペディアで「KernelICA」の詳細全文を読む スポンサード リンク 翻訳と辞書 : 翻訳のためのインターネットリソース Copyright(C) kotoba.ne.jp 1997-2016. All Rights Reserved.