Machine Learning Common subscript and their meanings

Copyright: Attribution, allow others to create paper-based, and must distribute paper (based on the original license agreement with the same license Creative Commons )

An asterisk * superscripts meaning

• Watermelon book: $(w^*,b^*)=\underset{(w,b)}{argmax}\sum_{i=1}^m(f(x_i)-y_i)^2$ meaning: that the expression takes a minimum value, the corresponding $w,b$ values.

• Li Hang "statistical learning methods": $d^*=\underset{\alpha,\beta;\alpha_i \geq 0 }{max}\theta_D(\alpha,\beta)$ $d^*$ Represents the optimal solution.

• Flower Book: usually a superscript * denotes minimized or maximized function $x$ values, such as in mind: $x^*=argminf(x)$

• That superscript asterisk * usually indicates the optimal solution

Tilde and the subscript meaning

Marginal distribution $P(X)$ empirical distribution $\tilde P(X)$ $\tilde P(X=x)=\frac{v(X=x)}{N}$

^ Broken line on the subject of meaning

• Li Hang "statistical learning methods": $R_{exp}(\hat y)=E_p[L(Y,\hat f(X))]$ , wherein $\hat f$Representation model learned
• Andrew Ng "deeplearning AI": $\hat y$The model indicates the predicted output vector

http://blog.csdn.net/u012965373/article/details/52936875

Continually updated