Home > Research > Publications & Outputs > Every team makes mistakes, in large action spaces

Electronic data

  • mpref15

    Accepted author manuscript, 463 KB, PDF document

Links

View graph of relations

Every team makes mistakes, in large action spaces

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Every team makes mistakes, in large action spaces. / Soriano Marcolino, Leandro; Nagarajan, Vaishnavh; Tambe, Milind.
9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015). 2015.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Soriano Marcolino, L, Nagarajan, V & Tambe, M 2015, Every team makes mistakes, in large action spaces. in 9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015). <http://ursaminor.informatik.uni-augsburg.de/mpref/mpref2015/download/every_team_makes_mistakes.pdf>

APA

Soriano Marcolino, L., Nagarajan, V., & Tambe, M. (2015). Every team makes mistakes, in large action spaces. In 9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015) http://ursaminor.informatik.uni-augsburg.de/mpref/mpref2015/download/every_team_makes_mistakes.pdf

Vancouver

Soriano Marcolino L, Nagarajan V, Tambe M. Every team makes mistakes, in large action spaces. In 9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015). 2015

Author

Soriano Marcolino, Leandro ; Nagarajan, Vaishnavh ; Tambe, Milind. / Every team makes mistakes, in large action spaces. 9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015). 2015.

Bibtex

@inproceedings{bfa33cb789f8485db12b681784d732f1,
title = "Every team makes mistakes, in large action spaces",
abstract = "Voting is applied to better estimate an optimal answer to complex problems in many domains. Werecently presented a novel benefit of voting, that has not been observed before: we can use the voting patterns to assess the performance of a team and predict whether it will be successful or not in problem-solving. Our prediction technique is completely domain independent, and it can be executed at any time during problem solving. In this paper we present a novel result about our technique: we show that the prediction quality increases with the size of the action space. We present a theoreticalexplanation for such phenomenon, and experiments in Computer Go with a variety of board sizes.",
author = "{Soriano Marcolino}, Leandro and Vaishnavh Nagarajan and Milind Tambe",
year = "2015",
language = "English",
booktitle = "9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015)",

}

RIS

TY - GEN

T1 - Every team makes mistakes, in large action spaces

AU - Soriano Marcolino, Leandro

AU - Nagarajan, Vaishnavh

AU - Tambe, Milind

PY - 2015

Y1 - 2015

N2 - Voting is applied to better estimate an optimal answer to complex problems in many domains. Werecently presented a novel benefit of voting, that has not been observed before: we can use the voting patterns to assess the performance of a team and predict whether it will be successful or not in problem-solving. Our prediction technique is completely domain independent, and it can be executed at any time during problem solving. In this paper we present a novel result about our technique: we show that the prediction quality increases with the size of the action space. We present a theoreticalexplanation for such phenomenon, and experiments in Computer Go with a variety of board sizes.

AB - Voting is applied to better estimate an optimal answer to complex problems in many domains. Werecently presented a novel benefit of voting, that has not been observed before: we can use the voting patterns to assess the performance of a team and predict whether it will be successful or not in problem-solving. Our prediction technique is completely domain independent, and it can be executed at any time during problem solving. In this paper we present a novel result about our technique: we show that the prediction quality increases with the size of the action space. We present a theoreticalexplanation for such phenomenon, and experiments in Computer Go with a variety of board sizes.

M3 - Conference contribution/Paper

BT - 9th Multidisciplinary Workshop on Advances in Preference Handling (M-PREF 2015)

ER -