2

I am looking for books on AI and Machine Learning written for PhD + level Mathematicians. Is there anything like this?

Thanks, Vladimir

Volodymyr Kruglov
  • 155
  • Maybe there is a unified approach to all those algorithms from the point of view of probabiliy theory/statistics when all machine learning are just special cases of a general theorem – Volodymyr Kruglov Jun 01 '16 at 13:59

1 Answers1

4

A lot of learning algorithms are basically empirical risk minimization.

You may want to look at "Understanding Machine Learning: From Theory to Algorithms" by Shalev-Shwartz and Ben-David, which is suitable for masters level students or Mohari's "Foundations of Machine Learning".

Devroye's A Probabilistic theory of Pattern Recognition is more intended for an applied mathematician audience (late undergrad/early grad) as is Vapniks' The Nature of Statistical Learning Theory. Books on concentration of measure like Lugosi, Massart and Bucheron's Concentration Inequaliaties text are useful too.

There are a bunch of graduate courses on statistical learning theory out there as well, such as this, this, this and others.

Batman
  • 19,390