Differences

This shows you the differences between two versions of the page.

Link to this comparison view

defensejoos [2014/09/25 16:47]
jbuijs
defensejoos [2014/10/14 17:03] (current)
jbuijs added talk title Barbara
Line 36: Line 36:
 10:00 - 10:05 opening 10:00 - 10:05 opening
  
-10:05 - 10:35 TBD+10:05 - 10:35 Metric learning and model interpretability (Barbara Hammer)
  
 10:35 - 11:05 Business Process Deviance Mining (Marlon Dumas) 10:35 - 11:05 Business Process Deviance Mining (Marlon Dumas)
  
-11:05 - 11:15 Pauze+11:05 - 11:15 Break
  
 11:15 - 11:45 Challenges in Information Visualization (Jack van Wijk) 11:15 - 11:45 Challenges in Information Visualization (Jack van Wijk)
Line 46: Line 46:
 11:45 - 12:15 An Overview of the Research Collaboration between TU/e and QUT in the area of Process Mining (Arthur ter Hofstede) 11:45 - 12:15 An Overview of the Research Collaboration between TU/e and QUT in the area of Process Mining (Arthur ter Hofstede)
  
 +
 +===== Abstract: Metric learning and model interpretability =====
 +
 +Metric learning aims at an automated adaptation of the distance measure 
 +which is used for the comparison of data points within a machine learning model.
 +It does not only greatly enhance the capability of popular distance-based
 +classifiers such as k-NN, clustering, or prototype based methods, but it also
 +facilitates model interpretability under the umbrella of so-called relevance
 +learning: the latter refers to the identification of the most popular features
 +and feature correlations for a given task at hand; notable
 +application range from biomedical data analysis up to the industrial process models.
 +Within the talk, I will give an overview about recent results on metric learning
 +for prototype based models. After a presentation of the general
 +principle of metric learning, its theoretical background and applications, I will focus on two recent research directions
 +which are of particular interest for applications:
 +How to guarantee valid model interpretability in the case of high data dimensionality?
 +How to link relevance learning to data visualisation?
 +
 +Presented by [[http://www.techfak.uni-bielefeld.de/~bhammer/|Barbara Hammer]]
  
  
 
defensejoos.txt · Last modified: 2014/10/14 17:03 by jbuijs     Back to top
Recent changes RSS feed Creative Commons License Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki Design by ChirripĆ³