Vol-1455
urn:nbn:de:0074-1455-8

Copyright © 2015 for the individual papers by the papers' authors. Copying permitted for private and academic purposes. This volume is published and copyrighted by its editors.

MetaSel 2015
Meta-Learning and Algorithm Selection

Proceedings of the 2015 International Workshop on Meta-Learning and Algorithm Selection
co-located with European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases 2015 (ECMLPKDD 2015)

Porto, Portugal, September 7th, 2015.

Edited by

Joaquin Vanschoren, Eindhoven University of Technology, The Netherlands
Pavel Brazdil, FEP, University of Porto / Inesc Tec, Portugal
Christophe Giraud-Carrier, Brigham Young University, USA
Lars Kotthoff, University of British Columbia, Canada

Table of Contents

Session 1: Invited Talks

Applying Model-Based Optimization to Hyperparameter Optimization in Machine Learning1-1
Bernd Bischl
On a Few Recent Developments in Meta-Learning for Algorithm Ranking and Selection2-2
Bernhard Pfahringer

Session 2: Full Papers

The Potential Benefits of Data Set Filtering and Learning Algorithm Hyperparameter Optimization3-14
Michael R. Smith, Tony Martinez, Christophe Giraud-Carrier
Learning Data Set Similarities for Hyperparameter Optimization Initializations15-26
Martin Wistuba, Nicolas Schilling, Lars Schmidt-Thieme
Limitations of Using Constraint Set Utility in Semi-Supervised Clustering27-42
Toon Van Craenendonck, Hendrik Blockeel
Dealing with Overlapping Clustering: A Constraint-based Approach to Algorithm Selection43-54
Antoine Adam, Hendrik Blockeel
Algorithm Selection via Meta-learning and Sample-based Active Testing55-66
Salisu M. Abdulrahman, Pavel Brazdil, Jan N. van Rijn, Joaquin Vanschoren
Metalearning for Multiple-Domain Transfer Learning67-79
Catarina Félix, Carlos Soares, Alípio Jorge
Meta-learning Recommendation of Default Hyper-parameter Values for SVMs in Classification Tasks80-92
Rafael G. Mantovani, André L. D. Rossi, Joaquin Vanschoren, André C.P.L.F. Carvalho
Sharing RapidMiner Workflows and Experiments with OpenML93-103
Jan N. van Rijn, Joaquin Vanschoren

Session 3: Extended Abstracts

Meta-QSAR: Learning How to Learn QSARs104-105
Ivan Olier, Crina Grosan, Noureddin Sadawi, Larisa Soldatova, Ross D. King
Concept of a Rule-based Configurator for Auto-WEKA Using OpenML106-107
Patryk Kiepas, Szymon Bobek, Grzegorz J. Nalepa
Generating Workflow Graphs Using Typed Genetic Programming108-109
Tomáš Kren, Martin Pilát, Klára Pešková, Roman Neruda
Preference-Based Meta-Learning using Dyad Ranking: Recommending Algorithms in Cold-Start Situations110-111
Dirk Schäfer, Eyke Hüllermeier
Towards a Collaborative Platform for Advanced Meta-Learning in Healthcare Predictive Analytics112-114
Milan Vukicevic, Sandro Radovanovic, Joaquin Vanschoren, Giulio Napolitano, Boris Delibasic
Study on Meta-Learning Approach Application in Rank Aggregation Algorithm Selection115-116
Alexey Zabashta, Ivan Smetannikov, Andrey Filchenkov

The whole proceedings can also be downloaded as a single file (PDF).

We offer a BibTeX file for citing papers of this workshop from LaTeX.

2015-09-22: submitted by Joaquin Vanschoren, metadata incl. bibliographic data published under Creative Commons CC0
2015-09-22: published on CEUR-WS.org |valid HTML5|