Home > Academics, Research, Reviews, Tutorials > Statistical Physics Algorithms in Machine Learning

Statistical Physics Algorithms in Machine Learning

Recently I let myself exposed to so many types of problems in machine learning that need pretty high-level techniques to deal with. I was so amazed that most of the methods are originated from Statistical Physics!!! The first one that I encounter is mean field algorithm which is like an invitation for me to other methods in statistical Physics such as Bethe free energy approximation, Kikuchi approximation, saddle point approximation, calculus of variations, etc. I guess there are a lot more and they are very useful for machine learning especially for graphical models framework in which the number of configurations can be intractable.

There are some materials that I found interesting

Constructing Free Energy Approximations and Generalized Belief Propagation Algorithms
by Jonathan S. Yedidia, William T. Freeman, and Yair Weiss

CCCP algorithms to minimize the Bethe and Kikuchi free energies: Convergent alternatives to belief propagation
by A. L. Yuille

Statistical Physics Algorithms That Converge
by A. L. Yuille and J. J. Kosowsky

Statistical Physics, Mixtures of Distributions, and the EM Algorithm
by Alan L. Yuille, Paul Stolorz, Joachim Utans

Statistical Physics of Clustering Algorithms
by Thore Graepel, Eckehard Sch, Prof Dr, Prof Dr, Klaus Obermayer

  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: