Re:The main trick in machine learning
by avaku on Monday December 09, 2013 @10:04PM
(#6878750)
Great explanation. I would like to add to this, that held-out data is often used in Bayesian learning too - for example, in cases when you intentionally over-specify the model (adding more parameters than might be needed) because you don't really know what the best model might be. The inference goes until the likelihood on held-out data keeps increasing. Example, gesture recognition in Kinekt. If someone finds this info useful, I also recommend Coursera course on Probabilistic Graphical Models.
[ Parent
| on HN ]
There are no comments on this comment yet.