Bishop's masterpiece "Pattern Recognition and machine learning" ready to read!

Bishop's masterpiece "Pattern recognitionand machine learning" has long been stationed in my hard drive for more than a year, Zennai fear of its vast number of pages, has not dared to start. Recently read the literature, repeatedly quoted. Had to turn it over and prepare to read it carefully. If you have the conditions, you should also write a reading note, or basically also look at the side and forget.

I shared the PDF on the V disk:

Http://vdisk.weibo.com/s/oM0W7

BISHOPDE Web page, here can download ppt and program:

http://research.microsoft.com/en-us/um/people/cmbishop/prml/

Watercress has a lot of good comments:

http://book.douban.com/subject/2061116/

About PRML

PRML is a classic in the field of pattern recognition and machine learning, published in 2007. The book author Christpher M. Bishop is the field of pattern recognition and machine learning, and its 1995 "Nerual Networks for pattern recognition" is also a classic in the field of pattern recognition and artificial neural networks.

PRML introduces the basic theory and main methods of pattern recognition and machine learning, and also covers some recent developments in the field of pattern recognition and machine learning, which is not only suitable for beginners, but also has great reference value for professional researchers.

A total of 738 pages, divided into 14 chapters, gradual, forward and backward echo, express clearly, understand deeply. Each chapter has corresponding exercises and answers, which is helpful for learning and teaching.

Book Review Reference

http://book.douban.com/review/4533178/

In fact, it took me almost two months to read it and not to say how much I understood it, but it did get a lot of work and a review. The introduction of the 1th chapter, not many said, after reading the book need to look back. The probability distribution of the 2nd chapter, written very well, although only a few simple distributions, but the concept of conjugate transcendental and exponential distribution of the introduction of the very clear, this chapter is the foundation of this book. The 3rd chapter and the 4th chapter of the linear classification and regression is a very good aspect is to use the Bayesisan point of view, should be to understand the basis of Baysian thought. The 5th chapter I did not read, directly skipped. (basically does not affect the later reading) The 6th chapter tells Guassian Process (this thing later I know is a non-parametric Bayessian method, now in the field of statistics is very popular. ) The 7th chapter is about SVM. The 8th chapter is the basis of modern graph-based model, need to read carefully, this chapter concept introduced very clear, many machine learning and computer vision paper now use the representation of the graph model can be explained here. The 9th Chapter EM algorithm, I think is a bright spot of the book, from the simplest k-mean, deduced Gaussian mixture model, and then to the promotion of EM algorithm, each section of this chapter is a boutique. In the 10th chapter, approximate inference is basically the basic principle of approximate inference in the first section and an example in section Ii. The method of Mean-field and Variational is adopted. The 11th chapter sampling, write very wonderful, to completely do not understand sampling of me, also can quickly get started. Here need to explain is, my harvest mainly from the 8th chapter to the 11th chapter, light reading is not, during, I mainly learn the most basic topic Model:lda. In the process of learning LDA, the 8th chapter to the 11th chapter is fully used. This feeling is very good. Recommend to everyone. The 12th chapter is PCA and some improvements, when used to see also time. The 13th chapter is the HMM model and the LDS, the two graph models are the same. Suggest to study hmm well, there should be other information for reference. The 14th chapter finally is the integration, many things now I do not understand very much. In short, this is a very good book, the key is to write a clear idea, focus on highlighting. It is recommended as a basic reference for reading papers.

Bishop's masterpiece "Pattern Recognition and machine learning" ready to read!