My way of entry, first from the application phenomenon, summed up the law to reverse the essence. It is wrong to plunge into theoretical books.
The foreigner's advanced, still manifests in the inheritance aspect. No way, we are under a lot of pressure, there are good things tucked away. Everybody's bitter.
I started from the introduction, with the basic concept, see XXX. Know the ten algorithms, you can start.
Open source components start, infer.net
The example is very classic, the explanation is fine, this is also the foreigner programmer to become quick the reason. Before looking at Libusb also so, programmer English is not good road is not far, I have deep experience.
Below is a brief introduction to the basic concepts, I am interested in the prediction aspect and so on machine classification learning.
Because people don't have to ask computers to be like people. Vision, hearing is not the ultimate right way. Computers have their own things, such as ultrasonic ranging instruments. Of course, visual and auditory is promising in terms of assisting the human side.
How does Bayesian understand. Search, about half an hour can be thoroughly understood.
Support Vector machines. Search
Http://www.cnblogs.com/jerrylead/archive/2011/03/13/1982639.html
And I like a lot of applications of wild programmers have resonance, even if 100 saw what I share a person has been harvested value.
When I was young, I wanted to make weapons so that everyone could have the power to change the world, grow up for so many years, and know that things are going to die fast, sharing weapons is power, and sharing knowledge is power.
With knowledge, not only open-minded, but also can produce money, money has a power. Do not envy capitalists and stars, each time has its own trend, I always believe that productivity/science is the driving force.
We cannot stand on the shoulders of giants to see farther, but we can follow the giant's path is also a kind of happiness.
Infer.net Open Source components: 1, Introduction to machine learning, from the Bayes.