Machine Learning Coffee seminar "Toward perfect density estimation"

Lecturer : 
Petri Myllymäki
Event type: 
HIIT seminar
Doctoral dissertation
Event time: 
2017-02-06 09:15 to 10:00
Konemiehentie 2, seminar room T5

The Machine Learning Coffee seminar (now featuring delicious porridge as well!) will continue on Monday, February 6th in Otaniemi.

Petri Myllymäki, Professor, Department of Computer Science, University of Helsinki


Toward Perfect Density Estimation

Abstract: We start by addressing a most simple problem, estimation of a one dimensional density function, and argue that despite of the apparent simplicity of the problem, it is surprisingly difficult to solve it in a holistic manner that is both computationally feasible and theoretically justifiable without strong distributional or other assumptions. We demonstrate how the information-theoretic MDL framework can be used for reaching this goal (almost) perfectly, and show how this simple setup gives interesting perspectives on the fundamental concepts in probabilistic modelling and statistical inference. We also discuss ideas for extending the framework to more complex models with additional practical applications.

Machine Learning Coffee seminars are weekly seminars held jointly by the Aalto University and the University of Helsinki. The seminars aim to gather people from different fields of science with interest in machine learning. Seminars will be held on Mondays at 9 am at Aalto University and the University of Helsinki every other week. At Aalto University, talks will be held in Konemiehentie 2, seminar room T5 and at the University of Helsinki in Kumpula, seminar room D123, unless otherwise noted. Talks will begin at 9:15 am and porridge and coffee will be served from 9:00 am.


Last updated on 2 Feb 2017 by Teemu Roos - Page created on 2 Feb 2017 by Teemu Roos