TY - GEN
T1 - Combinatorial online prediction via metarounding
AU - Fujita, Takahiro
AU - Hatano, Kohei
AU - Takimoto, Eiji
N1 - Copyright:
Copyright 2013 Elsevier B.V., All rights reserved.
PY - 2013
Y1 - 2013
N2 - We consider online prediction problems of combinatorial concepts. Examples of such concepts include s-t paths, permutations, truth assignments, set covers, and so on. The goal of the online prediction algorithm is to compete with the best fixed combinatorial concept in hindsight. A generic approach to this problem is to design an online prediction algorithm using the corresponding offline (approximation) algorithm as an oracle. The current state-of-the art method, however, is not efficient enough. In this paper we propose a more efficient online prediction algorithm when the offline approximation algorithm has a guarantee of the integrality gap.
AB - We consider online prediction problems of combinatorial concepts. Examples of such concepts include s-t paths, permutations, truth assignments, set covers, and so on. The goal of the online prediction algorithm is to compete with the best fixed combinatorial concept in hindsight. A generic approach to this problem is to design an online prediction algorithm using the corresponding offline (approximation) algorithm as an oracle. The current state-of-the art method, however, is not efficient enough. In this paper we propose a more efficient online prediction algorithm when the offline approximation algorithm has a guarantee of the integrality gap.
UR - http://www.scopus.com/inward/record.url?scp=84887443709&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84887443709&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-40935-6_6
DO - 10.1007/978-3-642-40935-6_6
M3 - Conference contribution
AN - SCOPUS:84887443709
SN - 9783642409349
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 68
EP - 82
BT - Algorithmic Learning Theory - 24th International Conference, ALT 2013, Proceedings
T2 - 24th International Conference on Algorithmic Learning Theory, ALT 2013
Y2 - 6 October 2013 through 9 October 2013
ER -