Details
| Original language | English |
|---|---|
| Title of host publication | Information Processing and Management of Uncertainty in Knowledge-Based Systems. Theory and Foundations - 17th International Conference, IPMU 2018, Proceedings |
| Editors | Jesus Medina, Bernadette Bouchon-Meunier, Ronald R. Yager, Jose Luis Verdegay, David A. Pelta, Manuel Ojeda-Aciego, Inma P. Cabrera |
| Publisher | Springer Verlag |
| Pages | 621-632 |
| Number of pages | 12 |
| ISBN (print) | 9783319914756 |
| Publication status | Published - 18 May 2018 |
| Externally published | Yes |
| Event | 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018 - Cadiz, Spain Duration: 11 Jun 2018 → 15 Jun 2018 |
Publication series
| Name | Communications in Computer and Information Science |
|---|---|
| Volume | 854 |
| ISSN (Print) | 1865-0929 |
Abstract
Games pose many interesting questions for the development of artificial intelligence agents. Especially popular are methods that guide the decision-making process of an autonomous agent, which is tasked to play a certain game. In previous studies, the heuristic search method Monte Carlo Tree Search (MCTS) was successfully applied to a wide range of games. Results showed that this method can often reach playing capabilities on par with humans or even better. However, the characteristics of collectible card games such as the online game Hearthstone make it infeasible to apply MCTS directly. Uncertainty in the opponent’s hand cards, the card draw, and random card effects considerably restrict the simulation depth of MCTS. We show that knowledge gathered from a database of human replays help to overcome this problem by predicting multiple card distributions. Those predictions can be used to increase the simulation depth of MCTS. For this purpose, we calculate bigram-rates of frequently co-occurring cards to predict multiple sets of hand cards for our opponent. Those predictions can be used to create an ensemble of MCTS agents, which work under the assumption of differing card distributions and perform simulations according to their assigned distribution. The proposed ensemble approach outperforms other agents on the game Hearthstone, including various types of MCTS. Our case study shows that uncertainty can be handled effectively using predictions of sufficient accuracy, ultimately, improving the MCTS guided decision-making process. The resulting decision-making based on such an MCTS ensemble proved to be less prone to errors by uncertainty and opens up a new class of MCTS algorithms.
Keywords
- Bigrams, Ensemble, Hearthstone, Monte Carlo Tree Search Knowledge-base, Uncertainty
ASJC Scopus subject areas
- Computer Science(all)
- General Computer Science
- Mathematics(all)
- General Mathematics
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
Information Processing and Management of Uncertainty in Knowledge-Based Systems. Theory and Foundations - 17th International Conference, IPMU 2018, Proceedings. ed. / Jesus Medina; Bernadette Bouchon-Meunier; Ronald R. Yager; Jose Luis Verdegay; David A. Pelta; Manuel Ojeda-Aciego; Inma P. Cabrera. Springer Verlag, 2018. p. 621-632 (Communications in Computer and Information Science; Vol. 854).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Predicting opponent moves for improving hearthstone AI
AU - Dockhorn, Alexander
AU - Frick, Max
AU - Akkaya, Ünal
AU - Kruse, Rudolf
PY - 2018/5/18
Y1 - 2018/5/18
N2 - Games pose many interesting questions for the development of artificial intelligence agents. Especially popular are methods that guide the decision-making process of an autonomous agent, which is tasked to play a certain game. In previous studies, the heuristic search method Monte Carlo Tree Search (MCTS) was successfully applied to a wide range of games. Results showed that this method can often reach playing capabilities on par with humans or even better. However, the characteristics of collectible card games such as the online game Hearthstone make it infeasible to apply MCTS directly. Uncertainty in the opponent’s hand cards, the card draw, and random card effects considerably restrict the simulation depth of MCTS. We show that knowledge gathered from a database of human replays help to overcome this problem by predicting multiple card distributions. Those predictions can be used to increase the simulation depth of MCTS. For this purpose, we calculate bigram-rates of frequently co-occurring cards to predict multiple sets of hand cards for our opponent. Those predictions can be used to create an ensemble of MCTS agents, which work under the assumption of differing card distributions and perform simulations according to their assigned distribution. The proposed ensemble approach outperforms other agents on the game Hearthstone, including various types of MCTS. Our case study shows that uncertainty can be handled effectively using predictions of sufficient accuracy, ultimately, improving the MCTS guided decision-making process. The resulting decision-making based on such an MCTS ensemble proved to be less prone to errors by uncertainty and opens up a new class of MCTS algorithms.
AB - Games pose many interesting questions for the development of artificial intelligence agents. Especially popular are methods that guide the decision-making process of an autonomous agent, which is tasked to play a certain game. In previous studies, the heuristic search method Monte Carlo Tree Search (MCTS) was successfully applied to a wide range of games. Results showed that this method can often reach playing capabilities on par with humans or even better. However, the characteristics of collectible card games such as the online game Hearthstone make it infeasible to apply MCTS directly. Uncertainty in the opponent’s hand cards, the card draw, and random card effects considerably restrict the simulation depth of MCTS. We show that knowledge gathered from a database of human replays help to overcome this problem by predicting multiple card distributions. Those predictions can be used to increase the simulation depth of MCTS. For this purpose, we calculate bigram-rates of frequently co-occurring cards to predict multiple sets of hand cards for our opponent. Those predictions can be used to create an ensemble of MCTS agents, which work under the assumption of differing card distributions and perform simulations according to their assigned distribution. The proposed ensemble approach outperforms other agents on the game Hearthstone, including various types of MCTS. Our case study shows that uncertainty can be handled effectively using predictions of sufficient accuracy, ultimately, improving the MCTS guided decision-making process. The resulting decision-making based on such an MCTS ensemble proved to be less prone to errors by uncertainty and opens up a new class of MCTS algorithms.
KW - Bigrams
KW - Ensemble
KW - Hearthstone
KW - Monte Carlo Tree Search Knowledge-base
KW - Uncertainty
UR - http://www.scopus.com/inward/record.url?scp=85056829819&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-91476-3_51
DO - 10.1007/978-3-319-91476-3_51
M3 - Conference contribution
AN - SCOPUS:85056829819
SN - 9783319914756
T3 - Communications in Computer and Information Science
SP - 621
EP - 632
BT - Information Processing and Management of Uncertainty in Knowledge-Based Systems. Theory and Foundations - 17th International Conference, IPMU 2018, Proceedings
A2 - Medina, Jesus
A2 - Bouchon-Meunier, Bernadette
A2 - Yager, Ronald R.
A2 - Verdegay, Jose Luis
A2 - Pelta, David A.
A2 - Ojeda-Aciego, Manuel
A2 - Cabrera, Inma P.
PB - Springer Verlag
T2 - 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018
Y2 - 11 June 2018 through 15 June 2018
ER -