Abstract
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.
Original language | English |
---|---|
Title of host publication | Algorithmic Learning Theory - 24th International Conference, ALT 2013, Proceedings |
Pages | 68-82 |
Number of pages | 15 |
DOIs | |
Publication status | Published - Nov 18 2013 |
Event | 24th International Conference on Algorithmic Learning Theory, ALT 2013 - Singapore, Singapore Duration: Oct 6 2013 → Oct 9 2013 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 8139 LNAI |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Other
Other | 24th International Conference on Algorithmic Learning Theory, ALT 2013 |
---|---|
Country | Singapore |
City | Singapore |
Period | 10/6/13 → 10/9/13 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Theoretical Computer Science
- Computer Science(all)
Cite this
Combinatorial online prediction via metarounding. / Fujita, Takahiro; Hatano, Kohei; Takimoto, Eiji.
Algorithmic Learning Theory - 24th International Conference, ALT 2013, Proceedings. 2013. p. 68-82 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8139 LNAI).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Combinatorial online prediction via metarounding
AU - Fujita, Takahiro
AU - Hatano, Kohei
AU - Takimoto, Eiji
PY - 2013/11/18
Y1 - 2013/11/18
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
ER -