Combining Parameter Space Search and Meta-Learning for Data-Dependent Computational Agent Recommendation

Autoři: Mgr. Roman Neruda, Ph.D.
Rok vydání: 2012
Typ publikace: Konferenční příspěvek (mezinárodní konf.)
Periodikum: Proceedings 2012 11th International Conference on Machine Learning and Applications ICMLA 2012
Projekty: Aproximace a učení funkcí mnoha proměnných pomocí neuronových sítí
Res Informatica
Bibliografická citace: KAZÍK, O., PEŠKOVÁ, K., PILÁT, M., NERUDA, Roman. Combining Parameter Space Search and Meta-Learning for Data-Dependent Computational Agent Recommendation. In: WANI, M.A., KHOSHGOFTAAR, T., ZHU, X., SELIYA, N., eds. Proceedings 2012 11th International Conference on Machine Learning and Applications ICMLA 2012. Los Alamitos: IEEE Computer Society, 2012, S. 36-41. ISBN 978-1-4673-4651-1
Dostupné z: doi:10.1109/ICMLA.2012.137
DOI: 10.1109/ICMLA.2012.137
ID: 384944

Abstrakt

The goal of our data-mining multi-agent system is to facilitate data-mining experiments without the necessary knowledge of the most suitable machine learning method and its parameters to the data. In order to replace the expert’s knowledge, the meta-learning subsystems are proposed including the parameter-space search and method recommendation based on previous experiments. In this paper we show the results of the parameter-space search with several search algorithms – tabulation, random search, simmulated annealing, and genetic algorithm.