Applying weak equivalence of categories between partial map and pointed set against changing the condition of 2-arms bandit problem

Takayuki Niizato, Yukio Gunji

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The main interest of the organizational learning is that the agent balances between exploring and exploiting. When the agent decides the answer in finite time, it gives rise to the problem that is trade-off between exploring and exploiting. This problem never arises in the given enough time. The agent must decide answer from his imperfect information, when the time is given finite for the task. 2-arms bandit problem is often taken into consideration in this problem. In previous article, we proposed the model for weak identity that is dynamically changing the hierarchy of his knowledge. Recently, the method of the heuristic model is attracted to approach this problem. In this article, we proposed different approaches for 2-arms bandit problem, and the model we proposed can adapt to the environment when the condition of the problem is changed and shows another approach to the organizational learning.

Original languageEnglish
Pages (from-to)10-21
Number of pages12
JournalComplexity
Volume16
Issue number4
DOIs
Publication statusPublished - 2011 Mar
Externally publishedYes

Fingerprint

Equivalence
Bandit problems
Organizational learning
Imperfect information
Trade-offs
Heuristics

Keywords

  • Adaptability
  • Exploration and exploitation
  • N-arms bandit problem
  • Organizational learning

ASJC Scopus subject areas

  • General

Cite this

Applying weak equivalence of categories between partial map and pointed set against changing the condition of 2-arms bandit problem. / Niizato, Takayuki; Gunji, Yukio.

In: Complexity, Vol. 16, No. 4, 03.2011, p. 10-21.

Research output: Contribution to journalArticle

@article{e755ff290d1c42318bd088712eabea54,
title = "Applying weak equivalence of categories between partial map and pointed set against changing the condition of 2-arms bandit problem",
abstract = "The main interest of the organizational learning is that the agent balances between exploring and exploiting. When the agent decides the answer in finite time, it gives rise to the problem that is trade-off between exploring and exploiting. This problem never arises in the given enough time. The agent must decide answer from his imperfect information, when the time is given finite for the task. 2-arms bandit problem is often taken into consideration in this problem. In previous article, we proposed the model for weak identity that is dynamically changing the hierarchy of his knowledge. Recently, the method of the heuristic model is attracted to approach this problem. In this article, we proposed different approaches for 2-arms bandit problem, and the model we proposed can adapt to the environment when the condition of the problem is changed and shows another approach to the organizational learning.",
keywords = "Adaptability, Exploration and exploitation, N-arms bandit problem, Organizational learning",
author = "Takayuki Niizato and Yukio Gunji",
year = "2011",
month = "3",
doi = "10.1002/cplx.20331",
language = "English",
volume = "16",
pages = "10--21",
journal = "Complexity",
issn = "1076-2787",
publisher = "John Wiley and Sons Inc.",
number = "4",

}

TY - JOUR

T1 - Applying weak equivalence of categories between partial map and pointed set against changing the condition of 2-arms bandit problem

AU - Niizato, Takayuki

AU - Gunji, Yukio

PY - 2011/3

Y1 - 2011/3

N2 - The main interest of the organizational learning is that the agent balances between exploring and exploiting. When the agent decides the answer in finite time, it gives rise to the problem that is trade-off between exploring and exploiting. This problem never arises in the given enough time. The agent must decide answer from his imperfect information, when the time is given finite for the task. 2-arms bandit problem is often taken into consideration in this problem. In previous article, we proposed the model for weak identity that is dynamically changing the hierarchy of his knowledge. Recently, the method of the heuristic model is attracted to approach this problem. In this article, we proposed different approaches for 2-arms bandit problem, and the model we proposed can adapt to the environment when the condition of the problem is changed and shows another approach to the organizational learning.

AB - The main interest of the organizational learning is that the agent balances between exploring and exploiting. When the agent decides the answer in finite time, it gives rise to the problem that is trade-off between exploring and exploiting. This problem never arises in the given enough time. The agent must decide answer from his imperfect information, when the time is given finite for the task. 2-arms bandit problem is often taken into consideration in this problem. In previous article, we proposed the model for weak identity that is dynamically changing the hierarchy of his knowledge. Recently, the method of the heuristic model is attracted to approach this problem. In this article, we proposed different approaches for 2-arms bandit problem, and the model we proposed can adapt to the environment when the condition of the problem is changed and shows another approach to the organizational learning.

KW - Adaptability

KW - Exploration and exploitation

KW - N-arms bandit problem

KW - Organizational learning

UR - http://www.scopus.com/inward/record.url?scp=79952741790&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952741790&partnerID=8YFLogxK

U2 - 10.1002/cplx.20331

DO - 10.1002/cplx.20331

M3 - Article

AN - SCOPUS:79952741790

VL - 16

SP - 10

EP - 21

JO - Complexity

JF - Complexity

SN - 1076-2787

IS - 4

ER -