The Many AI Challenges of Hearthstone

  • Amy K. HooverEmail author
  • Julian Togelius
  • Scott Lee
  • Fernando de Mesentier Silva
Technical Contribution


Since the inception of artificial intelligence, games have benchmarked algorithmic advances. Recent success in classic board games such as Chess and Go have left space for video games that pose related yet different sets of challenges. With this shifted focus, the set of AI problems associated with video games has expanded from simply playing these games to win, to include playing games in particular styles, generating game content, modeling players, etc. Different games pose different challenges for AI systems, and several such AI challenges can typically be addressed in the same game. In this article we analyze the popular collectible card game Hearthstone published by Blizzard in 2014, and describe a varied set of interesting AI challenges it poses. Despite their popularity and associated interesting challenges, collectible card games are relatively understudied in the AI community. By analyzing a single game in-depth, we get a glimpse of the entire field of AI and games through the lens of a single game, discovering a few new variations on existing research topics.


Artificial intelligence Games Hearthstone Deckbuilding Gameplaying Player modeling 



Thanks to Matthew Fontaine, Rodrigo Canaan, Aditya Bhatt, Connor Watson, Param Trivedi for their contributions to research that has informed this paper. Additionally, we thank Andy Nealen and Alex Zook for useful discussions. Finally, we are happy that all the hours we spent playing the game could contribute to something useful, or at least publishable.


  1. 1.
    Abrakam (2017) Faeria, AngleurGoogle Scholar
  2. 2.
    Agrawal R, Srikant R (1994) Fast algorithms for mining association rules. In: Proceedings of the 20th international conference on very large data bases, VLDB, vol 1215, pp 487–499Google Scholar
  3. 3.
    Aponte M-V, Levieux G, Natkin S (2009) Scaling the level of difficulty in single player video games. In: Natkin S, Dupire J (eds) Entertainment computing – ICEC 2009. ICEC 2009. Lecture notes in computer science, vol 5709. Springer, Berlin, pp 24–35Google Scholar
  4. 4.
    Bellemare MG, Naddaf Y, Veness J, Bowling M (2013) The arcade learning environment: an evaluation platform for general agents. J Artif Intell Res 47:253–279CrossRefGoogle Scholar
  5. 5.
    Bethesda Softworks (2017) The elder scrolls: legendsGoogle Scholar
  6. 6.
    Bhatt A, Lee S, de Mesentier Silva F, Watson CW, Togelius J, Hoover AK (2018) Exploring the Hearthstone deck space. In: Proceedings of the 13th international conference on the foundations of digital games (FDG). ACM, pp 18:1–18:10Google Scholar
  7. 7.
    Blizzard Entertainment (2014) Hearthstone, Irvine, CAGoogle Scholar
  8. 8.
    Browne C (2011) Evolutionary game design. SpringerBriefs in computer science, Springer, London, pp 75–85Google Scholar
  9. 9.
    Bursztein E (2016) I am a legend: hacking Hearthstone using statistical learning methods. In: Proceedings of the 2016 IEEE Conference on Computational Intelligence and Games (CIG), pp 1–8Google Scholar
  10. 10.
    Cai X, Wunsch DC (2007) Computer go: a grand challenge to AI. In: Challenges for computational intelligence. Springer, Berlin, pp 443–465Google Scholar
  11. 11.
    CD Projekt and CD Projekt RED (2019) Gwent: the witcher card gameGoogle Scholar
  12. 12.
    Coulom R (2006) Efficient selectivity and backup operators in Monte-Carlo tree search. In: Proceedings of the international conference on computers and games. Springer, pp 72–83Google Scholar
  13. 13.
    Cully A, Clune J, Tarapore D, Mouret J-B (2015) Robots that can adapt like animals. Nature 521(7553):503CrossRefGoogle Scholar
  14. 14.
    de Mesentier Silva F, Canaan R, Lee S, Togelius J, Hoover AK (2019) Evolving the hearthstone meta. In: Proceedings of the First IEEE Conference on Games (CoG)Google Scholar
  15. 15.
    de Mesentier Silva F, Isaksen A, Togelius J, Nealen A (2016) Generating heuristics for novice players. In: Proceedings of the 2016 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, pp 1–8Google Scholar
  16. 16.
    de Mesentier Silva F, Togelius J, Lantz F, Nealen A (2018) Generating beginner heuristics for simple Texas Hold’em. In: Proceedings of the 2018 genetic and evolutionary computation conference (GECCO). ACM, pp 181–188Google Scholar
  17. 17.
    de Mesentier Silva F, Togelius J, Lantz F, Nealen A (2018) Generating novice heuristics for post-flop poker. In: Proceedings of the 2018 IEEE conference on computational intelligence and games (CIG). IEEE, pp 1–8Google Scholar
  18. 18.
    Dockhorn A, Frick M, Akkaya Ü, Kruse R (2018) Predicting opponent moves for improving Hearthstone AI. In: Proceedings of the 18th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems. Springer, pp 621–632Google Scholar
  19. 19.
    Dire Wolf Digital LLC (2016) EternalGoogle Scholar
  20. 20.
    Drachen A, Sifa R, Bauckhage C, Thurau C (2012) Guns, swords and data: Clustering of player behavior in computer games in the wild. In: Proceedings of the 2012 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, pp 163–170Google Scholar
  21. 21.
    Ecoffet A, Huizinga J, Lehman J, Stanley KO, Clune J (2019) Go-explore: a new approach for hard-exploration problems. arXiv preprint arXiv:1901.10995
  22. 22.
    Elias GS, Garfield R, Gutschera KR (2012) Characteristics of games. MIT Press, CambridgeGoogle Scholar
  23. 23.
    Ensmenger N (2012) Is chess the drosophila of artificial intelligence? A social history of an algorithm. Soc Stud Sci 42(1):5–30CrossRefGoogle Scholar
  24. 24.
    Fontaine MC, Lee S, Soros LB, Silva FDM, Togelius J, Hoover AK (2019) Mapping Hearthstone deck spaces through MAP-Elites with sliding boundaries. In: Proceedings of the 2019 Genetic and Evolutionary Computation Conference (GECCO). ACMGoogle Scholar
  25. 25.
    García-Sánchez P, Tonda A, Squillero G, Mora A, Merelo JJ (2016) Evolutionary deckbuilding in Hearthstone. In: Proceedings of the 2016 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, pp 1–8Google Scholar
  26. 26.
    García-Sánchez P, Tonda A, Mora AM, Squillero G, Merelo JJ (2018) Automated playtesting in collectible card games using evolutionary algorithms: a case study in Hearthstone. Knowl Based Syst 153:133–146CrossRefGoogle Scholar
  27. 27.
    Gobet F, Simon HA (1996) The roles of recognition processes and look-ahead search in time-constrained expert problem solving: evidence from grand-master-level chess. Psychol Sci 7(1):52–55CrossRefGoogle Scholar
  28. 28.
    Góes LFW, Da Silva AR, Saffran J, Amorim A, França C, Zaidan T, Olímpio BMP, Alves LRO, Morais H, Luana S et al (2017) Honingstone: building creative combos with honing theory for a digital card game. IEEE Trans Comput Intell AI Games 9(2):204–209CrossRefGoogle Scholar
  29. 29.
    Green MC, Khalifa A, Barros GAB, Machado T, Nealen A, Togelius J (2018) AtDELFI: automatically designing legible, full instructions for games. In: Proceedings of the 13th international conference on the foundations of digital games (FDG). ACM, p 17Google Scholar
  30. 30.
    Guzdial M, Liao N, Riedl M (2018) Co-creative level design via machine learning. Fifth Experimental AI in Games Workshop (EXAG). arXiv preprint arXiv:1809.09420
  31. 31.
    Hodge V, Sephton N, Devlin S, Cowling P, Goumagias N, Shao J, Purvis K, Cabras I, Fernandes K, Li F (2019) How the business model of customisable card games influences player engagement. IEEE Trans Games.
  32. 32.
    Holmgard C, Green MC, Liapis A, Togelius J (2018) Automated playtesting with procedural personas with evolved heuristics. IEEE Trans Games.
  33. 33.
    Holmgård C, Liapis A, Togelius J, Yannakakis GN (2014) Evolving personas for player decision modeling. In: Proceedings of the 2014 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, pp 1–8Google Scholar
  34. 34.
    Hoover AK, Szerlip PA, Stanley KO (2014) Functional scaffolding for composing additional musical voices. Comput Music J 38(4):80–99CrossRefGoogle Scholar
  35. 35.
    Jakubik J (2018) A neural network approach to Hearthstone win rate prediction. In: Proceedings of the 2018 federated conference on computer science and information systems (FedCSIS). IEEE, pp 185–188Google Scholar
  36. 36.
    Janusz A, Świechowski M, Tajmajer T (2017) Helping AI to play Hearthstone: AAIA’17 data mining challenge. arXiv preprint arXiv:1708.00730
  37. 37.
    Janusz A, Tajmajer T, Świechowski M, Grad Ł, Puczniewski J, Ślȩzak D (2018) Toward an intelligent HS deck advisor: lessons learned from AAIA’18 data mining competition. In: Proceedings of the 2018 federated conference on computer science and information systems (FedCSIS). IEEE, pp 189–192Google Scholar
  38. 38.
    Kasparov G (2017) Deep thinking: where machine intelligence ends and human creativity begins. Public AffairsGoogle Scholar
  39. 39.
    Liapis A, Yannakakis GN, Togelius J (2013) Sentient sketchbook: computer-aided game level authoring. In: Proceedings of the 8th Conference on the Foundations of Digital Games (FDG), pp 213–220Google Scholar
  40. 40.
    Ling W, Grefenstette E, Hermann KM, Kočiskỳ T, Senior A, Wang F, Blunsom P (2016) Latent predictor networks for code generation. arXiv preprint arXiv:1603.06744
  41. 41.
    Lucas SM (2009) Computational intelligence and AI in games: a new IEEE transactions. IEEE Trans Comput Intell AI Games 1(1):1–3CrossRefGoogle Scholar
  42. 42.
    Machado T, Bravi I, Wang Z, Nealen A, Togelius J (2016) Shopping for game mechanics. In: Proceedings of the 7th workshop on procedural content generation the first joint conference of the joint international conference of DiGRA and FDGGoogle Scholar
  43. 43.
    Mahlmann T, Togelius J, Yannakakis GN (2012) Evolving card sets towards balancing dominion. In: Proceeding of the 2012 IEEE congress on evolutionary computation (CEC). IEEE, pp 1–8Google Scholar
  44. 44.
    Mnih V, Kavukcuoglu K, Silver D, Rusu AA, Veness J, Bellemare MG, Graves A, Riedmiller M, Fidjeland AK, Ostrovski G et al (2015) Human-level control through deep reinforcement learning. Nature 518(7540):529CrossRefGoogle Scholar
  45. 45.
    Mouret J-B, Clune J (2015) Illuminating search spaces by mapping elites. arXiv preprint arXiv:1504.04909
  46. 46.
    OpenAI (2018) OpenAI Five. Accessed 10 June 2019
  47. 47.
    Santos A, Santos PA, Melo FS (2017) Monte carlo tree search experiments in Hearthstone. In: Proceedings of the 2017 IEEE Conference on Computational Intelligence and Games (CIG). IEEE, pp 272–279Google Scholar
  48. 48.
    Shaker N, Shaker M, Togelius J (2013) Ropossum: an authoring tool for designing, optimizing and solving cut the rope levels. In: Proceedings of the ninth AAAI conference on artificial intelligence and interactive digital entertainmentGoogle Scholar
  49. 49.
    Shannon CE (1950) XXII. Programming a computer for playing chess. Lond Edinb Dublin Philos Mag J Sci 41(314):256–275CrossRefzbMATHGoogle Scholar
  50. 50.
    Silver D, Huang A, Maddison CJ, Guez A, Sifre L, Van Den Driessche G, Schrittwieser J, Antonoglou I, Panneershelvam V, Lanctot M et al (2016) Mastering the game of Go with deep neural networks and tree search. Nature 529(7587):484CrossRefGoogle Scholar
  51. 51.
    Smith G, Whitehead J, Mateas M (2011) Tanagra: Reactive planning and constraint solving for mixed-initiative level design. IEEE Trans Comput Intell AI Games 3(3):201–215CrossRefGoogle Scholar
  52. 52.
    Spronck P, Sprinkhuizen-Kuyper I, Postma E (2004) Difficulty scaling of game AI. In: Proceedings of the 5th International Conference on Intelligent Games and Simulation (GAME-ON 2004), pp 33–37Google Scholar
  53. 53.
    Stiegler A, Dahal K, Maucher J, Livingstone D (2017) Symbolic reasoning for Hearthstone. IEEE Trans Comput Intell AI Games 10(2):113-127CrossRefGoogle Scholar
  54. 54.
    Stiegler A, Messerschmidt C, Maucher J, Dahal K (2016) Hearthstone deck-construction with a Utility System. In: Proceedings of the 2016 10th international conference on software, knowledge, information management & applications (SKIMA). IEEE, pp 21–28Google Scholar
  55. 55.
    Summerville AJ, Mateas M (2016) Mystical tutor: a magic: the gathering design assistant via denoising sequence-to-sequence learning. In: Twelfth artificial intelligence and interactive digital entertainment conferenceGoogle Scholar
  56. 56.
    Supercell (2012) Clash of clansGoogle Scholar
  57. 57.
    Świechowski M, Tajmajer T, Janusz A (2018) Improving Hearthstone AI by combining MCTS and supervised learning algorithms. arXiv preprint arXiv:1808.04794
  58. 58.
    Turing AM (1953) Digital computers applied to games. Faster than Thought, PitmanGoogle Scholar
  59. 59.
    van Lankveld G, Spronck P, Rauterberg M (2008) Difficulty scaling through incongruity. In: Proceedings of the Fourth AAAI Conference on Artificial Intelligence and Interactive Digital EntertainmentGoogle Scholar
  60. 60.
    Vinyals O, Babuschkin I, Chung J, Mathieu M, Jaderberg M, Czarnecki WM, Dudzik A, Huang A, Georgiev P, Powell R, Ewalds T, Horgan D, Kroiss M, Danihelka I, Agapiou J, Oh J, Dalibard V, Choi D, Sifre L, Sulsky Y, Vezhnevets S, Molloy J, Cai T, Budden D, Paine T, Gulcehre C, Wang Z, Pfaff T, Pohlen T, Wu Y, Yogatama D, Cohen J, McKinney K, Smith O, Schaul T, Lillicrap T, Apps C, Kavukcuoglu K, Hassabis D, Silver D (2019) AlphaStar: mastering the real-time strategy game StarCraft II. Accessed 20 Apr 2019
  61. 61.
    Ward CD, Cowling PI (2009) Monte carlo search applied to card selection in magic: the gathering. In: 2009 IEEE symposium on computational intelligence and games (CIG). IEEE, pp 9–16Google Scholar
  62. 62.
    Wizards of the Coast (1993) Magic: the gathering, Renton, WashingtonGoogle Scholar
  63. 63.
    Yannakakis GN, Liapis A, Alexopoulos C (2014) Mixed-initiative co-creativity. In: Proceedings of the 9th Conference on the Foundations of Digital Games (FDG)Google Scholar
  64. 64.
    Yannakakis GN, Togelius J (2018) Artificial intelligence and games. Springer, BerlinCrossRefGoogle Scholar
  65. 65.
    Yee N (2006) Motivations for play in online games. CyberPsychol Behav 9(6):772–775CrossRefGoogle Scholar
  66. 66.
    Zhang S, Buro M (2017) Improving Hearthstone AI by learning high-level rollout policies and bucketing chance node events. In: Proceedings of the 2017 IEEE conference on computational intelligence and games (CIG). IEEE, pp 309–316Google Scholar
  67. 67.
    Zopf M (2015) A comparison between the usage of flat and structured game trees for move evaluation in Hearthstone. Master, Technische Universität Darmstadt, Darmstadt, GermanyGoogle Scholar

Copyright information

© Gesellschaft für Informatik e.V. and Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  • Amy K. Hoover
    • 1
    Email author
  • Julian Togelius
    • 2
  • Scott Lee
    • 3
  • Fernando de Mesentier Silva
    • 4
  1. 1.New Jersey Institute of TechnologyNewarkUSA
  2. 2.New York UniversityBrooklynUSA
  3. 3.IrvineUSA
  4. 4.Rio de JaneiroBrazil

Personalised recommendations