Action Selection for Hammer Shots in Curling: Optimization of Non-convex Continuous Actions With Stochastic Action OutcomesReport as inadecuate




Action Selection for Hammer Shots in Curling: Optimization of Non-convex Continuous Actions With Stochastic Action Outcomes - Download this document for free, or read online. Document in PDF available to download.

stochastic, optimization, curling, games

Ahmad, Zaheen F

Supervisor and department: Holte, Robert Computing Science

Examining committee member and department: Lewis, Mark Mathematical and Statistical Sciences Bowling, Michael Computing Science Holte, Robert Computing Science

Department: Department of Computing Science

Specialization:

Date accepted: 2017-01-20T13:42:05Z

Graduation date: 2017-06:Spring 2017

Degree: Master of Science

Degree level: Master's

Abstract: Optimal decision making in the face of uncertainty is an active area of research in artificial intelligence. In this thesis, I present the sport of curling as a novel application domain for research in optimal decision making. I focus on one aspect of the sport, the hammer shot, the last shot taken before a score is given, and how selecting this shot can be modelled as a low-dimensional optimization problem with a continuous action space and stochastic transitions. I explore the unique research challenges that are brought forth when optimizing in a setting where there is uncertainty in the action outcomes. I then survey several existing optimization strategies and describe a new optimization algorithm called Delaunay Sampling, adapted from a method based on Delaunay triangulation. I compare the performance of Delaunay Sampling with the other algorithms using our curling physics simulator and show that it outperforms these other algorithms. I also show that, with a few caveats, Delaunay Sampling exceeds the performance of Olympic-level humans when selecting strategies for hammer shots.

Language: English

DOI: doi:10.7939-R3MS3KD3C

Rights: This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for the purpose of private, scholarly or scientific research. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.





Author: Ahmad, Zaheen F

Source: https://era.library.ualberta.ca/


Teaser



Action Selection for Hammer Shots in Curling: Optimization of Non-convex Continuous Actions With Stochastic Action Outcomes by Zaheen Farraz Ahmad A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Computing Science University of Alberta © Zaheen Farraz Ahmad, 2017 Abstract Optimal decision making in the face of uncertainty is an active area of research in artificial intelligence.
In this thesis, I present the sport of curling as a novel application domain for research in optimal decision making.
I focus on one aspect of the sport, the hammer shot, the last shot taken before a score is given, and how selecting this shot can be modelled as a lowdimensional optimization problem with a continuous action space and stochastic transitions. I explore the unique research challenges that are brought forth when optimizing in a setting where there is uncertainty in the action outcomes.
I then survey several existing optimization strategies and describe a new optimization algorithm called Delaunay Sampling, adapted from a method based on Delaunay triangulation.
I compare the performance of Delaunay Sampling with the other algorithms using our curling physics simulator and show that it outperforms these other algorithms.
I also show that, with a few caveats, Delaunay Sampling exceeds the performance of Olympic-level humans when selecting strategies for hammer shots. ii To my family because otherwise my mom would never let me hear the end of it. iii Acknowledgements The utmost of gratitudes go to my supervisor Robert C.
Holte for his mentorship and guidance throughout the span of my degree.
I will always be thankful to him for providing me with the freedom to explore and to learn while also giving me direction when I needed it. I would also like to thank Michael Bowling for all the advice he gave me and for his work as a collaborator and on my committee.
I would also like express my gratitude to Mark Lewis...





Related documents