小白转DS，面试常见问题总结2，新人求大米

马上要到找工作季了，想要转ds，吐血总结一些面试中会问道的问题第二波。新人上路，求大米~
7,
Hyperparameter is usually tuned by cross validation method.

8, kernel function
I think Kernel function is a dot product of two vectors that represent map to other higher dimensional spaces. Using it save the effort to transform the data to a higher dimensional space and transform back. It is equivalent to that process.
Kernel is a way of computing the dot product of two vectors xx and yy in some (possibly very high dimensional) feature space. Kernels give a way to compute dot products in some feature space without even knowing what this space is and what is φ.
https://stats.stackexchange.com/ ... in-what-a-kernel-is
. Waral dи,
9. Precision and recall: all care about true positive
    Accuracy care all the predictions

10. ROC Curve
The receiver operating characteristic curve is plot which shows the performance of a binary classifier as function of its cut-off threshold
AUC (area under the curve) is an aggregated measure of performance of a binary classifier on all possible threshold values

11. Selection bias is a phenomenon that involves the selection of individual or grouped data in a way that is not considered to be random

12. Type one error: rejecting a true null hypothesis
Type two errorL accepting a false null hypothesis

如果大家觉得有用的话，会继续更新一些自己整理的常考知识点~真诚求米，感激~

How to avoid overfitting?
1, add regularization, L1 or L2.
2, add dropout layers.
3, early stop   4, proper test-validation and train data set split, avoid information leak
5, use a simpler model, 6, generate or collect more data

13, the moment-generating function of a real-valued random variable is an alternative specification of its probability distribution. Thus, it provides the basis of an alternative route to analytical results compared with working directly with probability density functions or cumulative distribution functions.
https://www.probabilitycourse.co ... oment_functions.php
14,  In statistics, the kth order statistic of a statistical sample is equal to its kth-smallest value.

15, for a positively skewed distribution please rank the mode mean and median.
The positively skewed distribution means a long tail in the positive axis, a peak in the negative axis, and thus make the mean>median>mode.