在C99里是不允许取register变量的地址的,见 6.5.3.2 Address and indirection operators Constraints 1 The operand of the unary & operator shall be either a function designator, the result of a [] or unary * operator, or an lvalue that designates an object that is not a bit-field and is not declared with the register storage-class specifier.
【在 t***s 的大作中提到】 : 这个人做的东西我比较了解。首先lssvm不是svm,两者的用的loss functioin : 不一样(square loss vs. hinge loss),所以lssvm toolbox跟libsvm或者 : svmlight不同,它不是一个svm solver。效果上,lssvm和svm不相上下,但用 : 的人并不多。这个人2000年左右搞出lssvm,发了非常多的paper,还写了一本书, : 但其实贡献并不大,因为lssvm不过是已经有的某些machine的新的提法而已 : (比如regularized networks,regularized kernel fda)。ml主流对他这种 : 鸡毛当令箭的做法不太看得上,甚至有人在文章里直接讽刺挖苦(e.g. ryan rifkin)。 : 他的publication里,icml 0, nips 1, jmlr 3,但都不是第一作者。结论: : 不在主流ml community里,本人学术水平一般。
t*s
26 楼
Everything Old is New Again: A Fresh Look at Historical Approaches in Machine Learning Ryan Rifkin, PhD Thesis, 2002 Page 100: Hans Suykens has also rederived the RLSC algorithm, under the name of Least Squares Support Vector Machines. He has been extremely prolific on the topic, publishing literally dozens of papers; the papers most relevant to the topic discussed in this thesis are [109,110,107,108,105,44,106]. However, these papers do not seem to advance either the theory or practice of RLSC. Like the work of Fung and Mangasarian, Suykens derives RLSC as a modification of Support Vector Machine, rather than directly as an implementation of Tikhonov regularization with a particular loss function, and seems somewhat ignorant of previous work in approximation theory. 是不是讽刺见仁见智,我是觉得已经夹枪带棒了。。
【在 t***s 的大作中提到】 : Everything Old is New Again: A Fresh Look at Historical : Approaches in Machine Learning : Ryan Rifkin, PhD Thesis, 2002 : Page 100: : Hans Suykens has also rederived the RLSC algorithm, under the : name of Least Squares Support Vector Machines. He has been : extremely prolific on the topic, publishing literally dozens of : papers; the papers most relevant to the topic discussed in this : thesis are [109,110,107,108,105,44,106]. However, these papers : do not seem to advance either the theory or practice of RLSC.
d*e
32 楼
写进Thesis真是有够mean
【在 t***s 的大作中提到】 : Everything Old is New Again: A Fresh Look at Historical : Approaches in Machine Learning : Ryan Rifkin, PhD Thesis, 2002 : Page 100: : Hans Suykens has also rederived the RLSC algorithm, under the : name of Least Squares Support Vector Machines. He has been : extremely prolific on the topic, publishing literally dozens of : papers; the papers most relevant to the topic discussed in this : thesis are [109,110,107,108,105,44,106]. However, these papers : do not seem to advance either the theory or practice of RLSC.