World Congress on Computational Intelligence (IEEE WCCI 2012), Brisbane, Australia
Special session: Meta-learning and hybrid systems for computational intelligence algorithms
Goals of this special session:
This session will be devoted to approaches that integrate in an intelligent way various learning algorithms used in computational intelligence, especially in meta-learning and hybrid systems. Integration of machine learning algorithms becomes increasingly more important, especially in applications to hard problems which still wait to be solved, where application of specialized methods that do not use additional knowledge has led to limited success. Data mining packages contain hundreds of algorithms that may be composed in millions of ways, and are able to beat the "no-free lunch" theorem, but automatization of this process requires analysis of learning algorithms at the meta-level. Methods that extract various forms of useful knowledge, share and integrate it for intelligent information processing, are necessary to solve hard problems. Such methods may be inspired by the organization of the brain, or may be based on formal algorithms. One promissing direction is to use methods that construct new features, learning from successes of divers algorithms, extracting knowledge from indirect, partial learning and using it to build final potential solutions. Another interesting aspect in construction of complex computational intelligence methods is dealing with different levels of abstraction; useful meta-knowledge may come in the form of highly abstract heuristic knowledge directing search process for optimal model, or may be hidden in details of algorithm implementation.
The main subjects of interest are:
· Meta-learning algorithms and system architectures.
· Meta-knowledge representation, acquisition, application, re-use and construction, analysis of the usefulness of knowledge.
· Hybrid systems, their architectures and ways of integration of individual models.
· Advanced architectures of data mining systems.
· Ensemble algorithms, advanced data mining systems/algorithms, which integrate several methods of data analysis.
· Deep learning algorithms.
· Transfer learning, information/knowledge sharing methods between modules of meta-learning or hybrid systems.
· Extraction and construction of new features that simplify the complex learning process, pre-processing methods, multimodal signal processing, extraction of information from specific types of data.
· Methods of reasoning for automatic creation of hybrid systems, estimation of knowledge usefulness for given problem.
· Methods of testing of complex systems, applications to challenging problems.
The session is not strictly limited to the above subjects. Every aspects of meta-learning or other integration of learning algorithms and knowledge are welcome.
Important dates:
Paper submission deadline | Dec 19, 2011 | ||
Paper acceptance notification date | Feb 20, 2012 | ||
Final paper submission deadline | April 2, 2012 | ||
(for more dates see http://www.ieee-wcci2012.org) |
Paper submission :
· Please REMEMBER to choose "Special session: Meta-learning and hybrid systems for computational intelligence algorithms"
Organizers:
Norbert Jankowski and Włodzisław Duch
Department of Informatics, Nicolaus Copernicus University
ul. Grudziadzka 5, 87-100 Toruń, Poland
{norbert,wduch} @ is.umk.pl
Program Committee:
Pavel Brazdil, Portugal, Andre C P L F de Carvalho, Brazil, Wlodzislaw Duch, Poland, Christophe Giraud-Carrier, USA, Krzysztof Grabczewski, Poland, Salman Halgamuge, Australia, Hisao Ishibuchi, Japan, Norbert Jankowski, Poland, Pavel Kordic, Czech Republic, Teresa Ludermir, Brazil, Ricardo Bastos C. Prudencio, Brazil, Rafał Scherer, Poland, Carlos Soares, Portugal, Bruno F. Souza, Brazil, Joaquin Vanschoren, Belgium, Wang Wenjia, Great Britain.
Web addresses:
Special session: http://www.is.umk.pl/WCCI2012-meta, WCCI'12: http://www.ieee-wcci2012.org
No comments:
Post a Comment