Logo PTI
Polish Information Processing Society
Logo FedCSIS

Annals of Computer Science and Information Systems, Volume 11

Proceedings of the 2017 Federated Conference on Computer Science and Information Systems

Multi-model approach for predicting the value function in the game of Heathstone: Heroes of Warcraft.

DOI: http://dx.doi.org/10.15439/2017F568

Citation: Proceedings of the 2017 Federated Conference on Computer Science and Information Systems, M. Ganzha, L. Maciaszek, M. Paprzycki (eds). ACSIS, Vol. 11, pages 139142 ()

Full text

Abstract. This document describes the problem presented at AAIA'17 Data Mining Challenge and my approach to solving it. In terms of reinforcement learning the task was to build an algorithm that predicts a value function for the game of Hearthstone: Heroes of Warcraft. I used an ensemble of 85 models trained on different features to build the final solution which scored the 36th place on the final leaderboard. Index Terms---data mining competition; classification; ranking; faeature engineering; algorithm composition;


  1. “Playing Atari With Deep Reinforcement Learning” Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra, Martin Riedmiller, NIPS Deep Learning Workshop, 2013.
  2. H. Cho, K. Kim, S. Cho, Replay-based Strategy Prediction and Build Order Adaptation for StarCraft AI Bots, IEEE CIG, 2013.
  3. M. Stanescu, S. Hernandez, G. Erickson, R. Greiner, M. Buro, Predicting Army Combat Outcomes in StarCraft, AAAI AIIDE, 2013.
  4. Y. N. Ravari, S. Bakkes, P. Spronck, StarCraft Winner Prediction, AAAI AIIDE, 2016.
  5. K. Conley and D. Perry, “How Does He Saw Me? A Recommendation Engine for Picking Heroes in Dota 2”, tech. rep., 2013.
  6. Kalyanaraman (2014). “To win or not to win? A prediction model to determine the outcome of a DotA2 match”. https://cseweb.ucsd.edu/~jmcauley/cse255/reports/wi15/Kaushik_Kalyanaraman.pdf
  7. Du, Xin, Jinjian Zhai, and Koupin Lv. “Algorithm Trading Using Q-Learning and Recurrent Reinforcement Learning.” CS229, n.d. Web. 15 Dec. 2016
  8. Yahya, A., Li, A., Kalakrishnan, M., Chebotar, Y., and Levine, S. (2016). Collective robot reinforcement learning with distributed asynchronous guided policy search. ArXiv e-prints
  9. Tom Fawcett. 2006. An introduction to ROC analysis. Pattern Recogn. Lett. 27, 8 (June 2006), 861-874. http://dx.doi.org/10.1016/j.patrec.2005.10.010
  10. Michael L. Littman. 2001. Value-function reinforcement learning in Markov games. Cogn. Syst. Res. 2, 1 (April 2001), 55-66. http://dx.doi.org/10.1016/S1389-0417(01)00015-8
  11. Game guide, http://us.battle.net/hearthstone/en/game-guide/
  12. Kaggle, https://www.kaggle.com/
  13. Kaggle Ensembling Guide, https://mlwave.com/kaggle-ensembling-guide/
  14. Breiman, L. Machine Learning (1996) 24: 123. http://dx.doi.org/10.1023/A:1018054314350
  15. Tianqi Chen and Carlos Guestrin. XGBoost: A Scalable Tree Boosting System. Preprint.
  16. Elie Bursztein, “I am a legend: Hacking Hearthstone using statistical learning methods“ https://cdn.elie.net/publications/i-am-a-legend-hacking-hearthstone-using-statistical-learning-methods.pdf
  17. Battle.net end user License Agreement http://us.blizzard.com/en-us/company/legal/eula.html