Online prediction under submodular constraints

Research output: Chapter in Book/Report/Conference proceedingConference contribution

15 Citations (Scopus)

Abstract

We consider an online prediction problem of combinatorial concepts where each combinatorial concept is represented as a vertex of a polyhedron described by a submodular function (base polyhedron). In general, there are exponentially many vertices in the base polyhedron. We propose polynomial time algorithms with regret bounds. In particular, for cardinality-based submodular functions, we give O(n 2)-time algorithms.

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings
Pages260-274
Number of pages15
Volume7568 LNAI
DOIs
Publication statusPublished - Oct 30 2012
Event23rd International Conference on Algorithmic Learning Theory, ALT 2012 - Lyon, France
Duration: Oct 29 2012Oct 31 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7568 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other23rd International Conference on Algorithmic Learning Theory, ALT 2012
CountryFrance
CityLyon
Period10/29/1210/31/12

Fingerprint

Polyhedron
Submodular Function
Prediction
Regret
Polynomials
Polynomial-time Algorithm
Cardinality
Vertex of a graph
Concepts

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Suehiro, D., hatano, K., Kijima, S., Takimoto, E., & Nagano, K. (2012). Online prediction under submodular constraints. In Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings (Vol. 7568 LNAI, pp. 260-274). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7568 LNAI). https://doi.org/10.1007/978-3-642-34106-9_22

Online prediction under submodular constraints. / Suehiro, Daiki; hatano, kohei; Kijima, Shuji; Takimoto, Eiji; Nagano, Kiyohito.

Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings. Vol. 7568 LNAI 2012. p. 260-274 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7568 LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Suehiro, D, hatano, K, Kijima, S, Takimoto, E & Nagano, K 2012, Online prediction under submodular constraints. in Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings. vol. 7568 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7568 LNAI, pp. 260-274, 23rd International Conference on Algorithmic Learning Theory, ALT 2012, Lyon, France, 10/29/12. https://doi.org/10.1007/978-3-642-34106-9_22
Suehiro D, hatano K, Kijima S, Takimoto E, Nagano K. Online prediction under submodular constraints. In Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings. Vol. 7568 LNAI. 2012. p. 260-274. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-34106-9_22
Suehiro, Daiki ; hatano, kohei ; Kijima, Shuji ; Takimoto, Eiji ; Nagano, Kiyohito. / Online prediction under submodular constraints. Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings. Vol. 7568 LNAI 2012. pp. 260-274 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{a9e9b9c352654f05a5d3e9bb3c75c6df,
title = "Online prediction under submodular constraints",
abstract = "We consider an online prediction problem of combinatorial concepts where each combinatorial concept is represented as a vertex of a polyhedron described by a submodular function (base polyhedron). In general, there are exponentially many vertices in the base polyhedron. We propose polynomial time algorithms with regret bounds. In particular, for cardinality-based submodular functions, we give O(n 2)-time algorithms.",
author = "Daiki Suehiro and kohei hatano and Shuji Kijima and Eiji Takimoto and Kiyohito Nagano",
year = "2012",
month = "10",
day = "30",
doi = "10.1007/978-3-642-34106-9_22",
language = "English",
isbn = "9783642341052",
volume = "7568 LNAI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "260--274",
booktitle = "Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings",

}

TY - GEN

T1 - Online prediction under submodular constraints

AU - Suehiro, Daiki

AU - hatano, kohei

AU - Kijima, Shuji

AU - Takimoto, Eiji

AU - Nagano, Kiyohito

PY - 2012/10/30

Y1 - 2012/10/30

N2 - We consider an online prediction problem of combinatorial concepts where each combinatorial concept is represented as a vertex of a polyhedron described by a submodular function (base polyhedron). In general, there are exponentially many vertices in the base polyhedron. We propose polynomial time algorithms with regret bounds. In particular, for cardinality-based submodular functions, we give O(n 2)-time algorithms.

AB - We consider an online prediction problem of combinatorial concepts where each combinatorial concept is represented as a vertex of a polyhedron described by a submodular function (base polyhedron). In general, there are exponentially many vertices in the base polyhedron. We propose polynomial time algorithms with regret bounds. In particular, for cardinality-based submodular functions, we give O(n 2)-time algorithms.

UR - http://www.scopus.com/inward/record.url?scp=84867856605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867856605&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-34106-9_22

DO - 10.1007/978-3-642-34106-9_22

M3 - Conference contribution

AN - SCOPUS:84867856605

SN - 9783642341052

VL - 7568 LNAI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 260

EP - 274

BT - Algorithmic Learning Theory - 23rd International Conference, ALT 2012, Proceedings

ER -