Introduction to Pattern recognition and machine learning

Pattern recognition originated in engineering, and machine learning originated in computer science. However, these different disciplines can be seen as a different direction in a field and have experienced considerable development over the last few decades. It is particularly pointed out that the Bayesian method (Bayesian methods) has changed from the patented method of the past (specialist niche) to the mainstream method (mainstream), and the graph model is used as a general framework to describe and apply probabilistic models. The practical practicability of the Bayesian method greatly facilitates the development of a series of approximate inference algorithms (approximate inference algorithms), such as variable decibel Dean (variational Bayes) and desired propagation (expectation Propagation). Similarly, the new kernels-based model has had a significant impact on both algorithms and practical applications.

The above is a brief overview of machine learning, in the future will be published a series of pattern recognition and machine learning articles, I hope to help you more, please criticize.

These articles are part of the translation and learning notes of the "Pattern Recognition and machine learning" book, and if there is a copyright issue, the copyright belongs to the original author.