This volume contains the papers presented at the 25th International
Conference on Algorithmic Learning Theory (ALT 2014), which was held
in Bled, Slovenia, from October 8–10, 2014.
ALT 2014 was co-located with
the 17th International Conference on Discovery Science (DS 2014).
The technical program of ALT 2014 had 4 invited talks
(presented jointly to both ALT 2014 and DS 2014) and 21 papers selected
from 50 submissions by the ALT Program Committee.
ALT 2014 took place in the hotel Golf in a beautiful park full of old trees in the very heart of Bled. It provided a stimulating interdisciplinary forum to discuss the theoretical foundations of machine learning as well as their relevance to practical applications.
ALT is dedicated to the theoretical foundations of machine learning and provides a forum for high-quality talks and scientific interaction in areas such as reinforcement learning, inductive inference and grammatical inference, learning from queries, active learning, probably approximate correct learning, online learning, bandit theory, statistical learning theory, Bayesian and stochastic learning, unsupervised or semi-supervised learning, clustering, universal prediction, stochastic optimization, high dimensional and non-parametric inference, information-based methods, decision tree methods, kernel-based methods, graph methods and/or manifold-based methods, sample complexity, complexity of learning, privacy preserving learning, learning based on Kolmogorov complexity, new learning models and applications of algorithmic learning theory.
The present volume of LNAI contains the text of the 21 papers presented at ALT 2014, as well as the texts/abstracts of the invited talks:
ALT 2014 was the 25th meeting in the ALT conference series, established in Japan in 1990. The ALT series is supervised by its Steering Committee: Peter Auer (University of Leoben, Austria), Shai Ben-David (University of Waterloo, Canada), Nader H. Bshouty (Technion - Israel Institute of Technology, Israel), Alexander Clark (King's College London, UK), Marcus Hutter (Australian National University, Canberra, Australia), Jyrki Kivinen (University of Helsinki, Finland), Frank Stephan (National University of Singapore, Republic of Singapore), Gilles Stoltz (Ecole normale supérieure, Paris, France), Csaba Szepesvári (University of Alberta, Edmonton, Canada), Eiji Takimoto (Kyushu University, Fukuoka, Japan), György Turán (University of Illinois at Chicago, USA, and University of Szeged, Hungary), Akihiro Yamamoto (Kyoto University, Japan), Thomas Zeugmann (Chair, Hokkaido University, Sapporo, Japan), and Sandra Zilles (co-Chair, University of Regina, Saskatchewan, Canada).
We thank various people and institutions who contributed to the success of the conference. Most importantly, we would like to thank the authors for contributing and presenting their work at the conference. Without their contribution this conference would not have been possible. We would like to thank the Office of Naval Research Global for the generous financial support for the conference ALT 2014 provided under ONRG GRANT N62909-14-1-C195.
ALT 2014 and DS 2014 were organized by the Jožef Stefan Institute (JSI) and the University of Ljubljana. We are very grateful to the Department of Knowledge Technologies (and the project MAESTRA) at JSI for sponsoring the conferences and providing administrative support. In particular, we thank the local arrangement chair, Mili Bauer, and her team, Tina Anžič, Nikola Simidjievski, and Jurica Levatić from JSI for their efforts in organizing the two conferences.
We are grateful for the collaboration with the conference series Discovery Science. In particular we would like to thank the General Chair of DS 2014 and ALT 2014 Ljupčo Todorovski and the DS 2014 Program Committee Chairs Sašo Džeroski, Dragi Kocev, and Panče Panov.
We are also grateful to EasyChair, the excellent conference management system, which was used for putting together the program for ALT 2014. EasyChair was developed mainly by Andrei Voronkov and is hosted at the University of Manchester. The system is cost-free.
We are grateful to the members of the Program Committee for ALT 2014 and the subreferees for their hard work in selecting a good program for ALT 2014. Last but not the least, we thank Springer for their support in preparing and publishing this volume in the Lecture Notes in Artificial Intelligence series.
General Chair for ALT 2014 and DS 2014
ALT 2014 Local Arrangements
back to the ALT Archives