这个有点太浅了,没任何证明,适合非 CS 本科生。不适合数理背景的人。 我跟了一个学期,觉得有点浪费时间。 这方面除了 Bishop 以外, Duda Hart Stork 也是比较经典的,然后就是去看 micheal jordan 的 paper 了。
【在 e***n 的大作中提到】 : www.ml-class.org
w*y
8 楼
Actually if you google statistical learning, the first one is the bible of statistical learning in my opinion. http://www-stat.stanford.edu/~tibs/ElemStatLearn/ If you have elemental statistical background, I will strongly recommend this one. Hastie's background is not that academical, so the book is easy to read instead of theoretically opaque.