翻訳と辞書 Words near each other ・ Hinlopen ・ Hinlopenbreen ・ Hinlopenstretet ・ Hinman ・ Hinman collator ・ Hinman Glacier ・ Hinman Settler Road ・ Hinmansville, New York ・ Hinn ・ Hinna ・ Hinnaard ・ Hinge (surname) ・ Hinge and Bracket ・ Hinge joint ・ Hinge line ・ Hinge loss ・ Hinge Studios ・ Hinge teeth ・ Hinge theorem ・ Hinged dissection ・ Hinged expansion joint ・ Hingede öö ・ Hingelind ・ Hingemouth ・ Hingene Castle ・ Hingert ・ Hinges ・ Hinges (album) ・ Hinges, Pas-de-Calais ・ Hinggan 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 Hinge loss ： ウィキペディア英語版 Hinge loss In machine learning, the hinge loss is a loss function used for training classifiers. The hinge loss is used for "maximum-margin" classification, most notably for support vector machines (SVMs). For an intended output and a classifier score , the hinge loss of the prediction is defined as :$\backslash ell(y)\; =\; \backslash max(0,\; 1-t\; \backslash cdot\; y)$ Note that should be the "raw" output of the classifier's decision function, not the predicted class label. For instance, in linear SVMs, $y\; =\; \backslash mathbf\; \backslash cdot\; \backslash mathbf\; +\; b$, where $(\backslash mathbf,b)$ are the parameters of the hyperplane and $\backslash mathbf$ is the point to classify. It can be seen that when and have the same sign (meaning predicts the right class) and $|y|\; \backslash ge\; 1$, the hinge loss $\backslash ell(y)\; =\; 0$, but when they have opposite sign, $\backslash ell(y)$ increases linearly with (one-sided error). ==Extensions== While SVMs are commonly extended to multiclass classification in a one-vs.-all or one-vs.-one fashion, there exists a "true" multiclass version of the hinge loss due to Crammer and Singer, defined for a linear classifier as :$\backslash ell(y)\; =\; \backslash max(0,\; 1\; +\; \backslash max\_\; \backslash mathbf\_t\; \backslash mathbf\; -\; \backslash mathbf\_y\; \backslash mathbf)$ In structured prediction, the hinge loss can be further extended to structured output spaces. Structured SVMs with margin rescaling use the following variant, where denotes the SVM's parameters, the joint feature function, and the Hamming loss: :$\backslash begin$ \ell(\mathbf) & = \max(0, \Delta(\mathbf, \mathbf) + \langle \mathbf, \phi(\mathbf, \mathbf) \rangle - \langle \mathbf, \phi(\mathbf, \mathbf) \rangle) \\ & = \max(0, \max_, \mathbf) + \langle \mathbf, \phi(\mathbf, \mathbf) \rangle \right) - \langle \mathbf, \phi(\mathbf, \mathbf) \rangle) \end 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「Hinge loss」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.