PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Three Paths to Relevance
Samuel Kaski
In: Brain-Inspired Information Technology (2010) Springer , Berlin , pp. 11-13.

Abstract

The problem of distinguishing between relevant and irrelevant variation is shared by natural and artificial data analysis systems. The problem is especially hard when the dimensionality of the data is high and yet inferences need to be made based on only few samples, and when the models are flexible machine learning models. I will discuss three learning strategies for coping with the problem, and applications in brain imaging, bioinformatics, and information retrieval.

EPrint Type:Book Section
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:6285
Deposited By:Samuel Kaski
Deposited On:08 March 2010