TY - GEN
T1 - Player Modeling via Multi-Armed Bandits
AU - Gray, Robert C.
AU - Zhu, Jichen
AU - Arigo, Danielle
AU - Forman, Evan
AU - Ontañón, Santiago
N1 - Funding Information:
This work is partially supported by the National Science Foundation under Grant Number IIS-1816470. The authors would like to thank the participants of our user study and all current and past members of this project. Special thanks to Jennifer Villareale and Diane Dallal for facilitating the user study and user data collection which is used in this paper.
Publisher Copyright:
© 2020 ACM.
PY - 2020/9/15
Y1 - 2020/9/15
N2 - This paper focuses on building personalized player models solely from player behavior in the context of adaptive games. We present two main contributions: The first is a novel approach to player modeling based on multi-armed bandits (MABs). This approach addresses, at the same time and in a principled way, both the problem of collecting data to model the characteristics of interest for the current player and the problem of adapting the interactive experience based on this model. Second, we present an approach to evaluating and fine-tuning these algorithms prior to generating data in a user study. This is an important problem, because conducting user studies is an expensive and labor-intensive process; therefore, an ability to evaluate the algorithms beforehand can save a significant amount of resources. We evaluate our approach in the context of modeling players' social comparison orientation (SCO) and present empirical results from both simulations and real players.
AB - This paper focuses on building personalized player models solely from player behavior in the context of adaptive games. We present two main contributions: The first is a novel approach to player modeling based on multi-armed bandits (MABs). This approach addresses, at the same time and in a principled way, both the problem of collecting data to model the characteristics of interest for the current player and the problem of adapting the interactive experience based on this model. Second, we present an approach to evaluating and fine-tuning these algorithms prior to generating data in a user study. This is an important problem, because conducting user studies is an expensive and labor-intensive process; therefore, an ability to evaluate the algorithms beforehand can save a significant amount of resources. We evaluate our approach in the context of modeling players' social comparison orientation (SCO) and present empirical results from both simulations and real players.
UR - http://www.scopus.com/inward/record.url?scp=85092322830&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092322830&partnerID=8YFLogxK
U2 - 10.1145/3402942.3402952
DO - 10.1145/3402942.3402952
M3 - Conference contribution
AN - SCOPUS:85092322830
T3 - ACM International Conference Proceeding Series
BT - Proceedings of the 15th International Conference on the Foundations of Digital Games, FDG 2020
A2 - Yannakakis, Georgios N.
A2 - Liapis, Antonios
A2 - Penny, Kyburz
A2 - Volz, Vanessa
A2 - Khosmood, Foaad
A2 - Lopes, Phil
PB - Association for Computing Machinery
T2 - 15th International Conference on the Foundations of Digital Games, FDG 2020
Y2 - 15 September 2020 through 18 September 2020
ER -