English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse
  1. Item Summary / 
  2. View item

Item

ITEM ACTIONSEXPORT
Add to Basket
 Local TagsRelease HistoryDetailsSummary
  Learning and decisions in contextual multi-armed bandit tasks

Schulz, E., Konstantinidis, E., & Speekenbrink, M. (2015). Learning and decisions in contextual multi-armed bandit tasks. In D. Noelle, R. Dale, A. Warlaumont, J. Yoshimi, T. Matlock, C. Jennings, et al. (Eds.), 37th Annual Meeting of the Cognitive Science Society (CogSci 2015): Mind, Technology and Society (pp. 2122-2127). Austin, TX, USA: Cognitive Science Society.

Item is

 

show all hide all

Basic

show hide
Item Permalink: https://hdl.handle.net/21.11116/0000-0006-B43A-E Version Permalink: https://hdl.handle.net/21.11116/0000-0006-B43B-D
Genre: Conference Paper

Files

show Files

Locators

show
hide
Locator:
https://cogsci.mindmodeling.org/2015/papers/0366/paper0366.pdf (Publisher version)
Description:
-
OA-Status:

Creators

show
hide
 Creators:
Schulz, E1, Author           
Konstantinidis, E, Author
Speekenbrink, M, Author
Affiliations:
1External Organizations, ou_persistent22              

Content

show
hide
Free keywords: -
 Abstract: Contextual Multi-Armed Bandit (CMAB) tasks are a novel framework to assess decision making in uncertain environments. In a CMAB task, participants are presented with multiple options (arms) which are characterized by a number of features (context) related to the reward associated with the arms. By choosing arms repeatedly and observing the reward, participants can learn about the relation between context and reward and improve their decision strategy. We present two studies on how people behave in CMAB tasks. Within a stationary environment, we find that participants are best described by Thompson Sampling-based Gaussian Process models. In a dynamic CMAB task we again find that participants are best described by probability matching of Gaussian Process expectations. Our findings imply that behavior previously referred to as "irrational" can actually be seen as a well-adapted strategy based on powerful inference algorithms.

Details

show
hide
Language(s):
 Dates: 2015-072015-11
 Publication Status: Issued
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: -
 Degree: -

Event

show
hide
Title: 37th Annual Meeting of the Cognitive Science Society (CogSci 2015)
Place of Event: Pasadena, CA, USA
Start-/End Date: 2015-07-22 - 2015-07-25

Legal Case

show

Project information

show

Source 1

show
hide
Title: 37th Annual Meeting of the Cognitive Science Society (CogSci 2015): Mind, Technology and Society
Source Genre: Proceedings
 Creator(s):
Noelle, DC, Editor
Dale, R, Editor
Warlaumont, AS, Editor
Yoshimi, J, Editor
Matlock, T, Editor
Jennings, CD, Editor
Maglio, PP, Editor
Affiliations:
-
Publ. Info: Austin, TX, USA : Cognitive Science Society
Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 2122 - 2127 Identifier: ISBN: 978-0-9911967-2-2