Approximate inference

src: slideplayer.com

Approximate inference methods make it possible to learn realistic models from big data by trading off computation time for accuracy, when exact learning and inference are computationally intractable.

Video Approximate inference

Major methods classes

• Variational Bayesian methods
• Expectation propagation
• Markov random fields
• Bayesian networks
  • Variational message passing
• loopy and generalized belief propagation

Maps Approximate inference

See also

• Statistical inference
• fuzzy logic
• data mining

src: pbs.twimg.com

References

src: xianblog.files.wordpress.com

External links

• Tom Minka, Microsoft Research (Nov 2, 2009). "Machine Learning Summer School (MLSS), Cambridge 2009, Approximate Inference" (video lecture). 

Source of the article : Wikipedia