Predicting Game Difficulty and Churn Without Players

We propose a novel simulation model that is able to predict the per-level churn and pass rates of Angry Birds Dream Blast, a popular mobile free-to-play game. Our primary contribution is to combine AI gameplay using Deep Reinforcement Learning (DRL) with a simulation of how the player population evolves over the levels. The AI players predict level difficulty, which is used to drive a player population model with simulated skill, persistence, and boredom. This allows us to model, e.g., how less persistent and skilled players are more sensitive to high difficulty, and how such players churn early, which makes the player population and the relation between difficulty and churn evolve level by level. Our work demonstrates that player behavior predictions produced by DRL gameplay can be significantly improved by even a very simple population-level simulation of individual player differences, without requiring costly retraining of agents or collecting new DRL gameplay data for each simulated player.

[1]  David Reguera,et al.  The Physics of Fun: Quantifying Human Engagement into Playful Activities , 2019, 1911.01864.

[2]  Aysu Betin Can,et al.  Automated Video Game Testing Using Synthetic and Humanlike Agents , 2019, IEEE Transactions on Games.

[3]  Kennon M. Sheldon,et al.  What is satisfying about satisfying events? Testing 10 candidate psychological needs. , 2001, Journal of personality and social psychology.

[4]  Kenneth R. Koedinger,et al.  Optimizing challenge in an educational game using large-scale design experiments , 2013, CHI.

[5]  Jari Takatalo,et al.  Presence, Involvement, and Flow in Digital Games , 2010, Evaluating User Experience in Games.

[6]  Dongbin Zhao,et al.  A Survey of Deep Reinforcement Learning in Video Games , 2019, ArXiv.

[7]  Alexei A. Efros,et al.  Curiosity-Driven Exploration by Self-Supervised Prediction , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[8]  Guillaume Lample,et al.  Playing FPS Games with Deep Reinforcement Learning , 2016, AAAI.

[9]  日経BP社,et al.  Amazon Web Services完全ソリューションガイド , 2016 .

[10]  Alex Graves,et al.  Playing Atari with Deep Reinforcement Learning , 2013, ArXiv.

[11]  Tapio Pahikkala,et al.  Playtime Measurement With Survival Analysis , 2017, IEEE Transactions on Games.

[12]  Tom Schaul,et al.  StarCraft II: A New Challenge for Reinforcement Learning , 2017, ArXiv.

[13]  Henryk Michalewski,et al.  Distributed Deep Reinforcement Learning: Learn how to play Atari games in 21 minutes , 2018, ISC.

[14]  Mohsen Sardari,et al.  Winning Is Not Everything: Enhancing Game Development With Intelligent Agents , 2019, IEEE Transactions on Games.

[15]  Jukka Heikkonen,et al.  Modelling user retention in mobile games , 2016, 2016 IEEE Conference on Computational Intelligence and Games (CIG).

[16]  Amitai Shenhav,et al.  The Eighty Five Percent Rule for optimal learning , 2018, Nature Communications.

[17]  Julian Togelius,et al.  AI-based playtesting of contemporary board games , 2017, FDG.

[18]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[19]  P. Silvia Interest—The Curious Emotion , 2008 .

[20]  Andrew K. Przybylski,et al.  The Motivational Pull of Video Games: A Self-Determination Theory Approach , 2006 .

[21]  Nikolaus Hansen,et al.  The CMA Evolution Strategy: A Tutorial , 2016, ArXiv.

[22]  Jeroen Jansz,et al.  Exploring persistence in gaming: The role of self-determination and social identity , 2014, Comput. Hum. Behav..

[23]  Ildar Kamaldinov,et al.  Deep Reinforcement Learning in Match-3 Game , 2019, 2019 IEEE Conference on Games (CoG).

[24]  Julian Togelius,et al.  Modeling Player Experience for Content Creation , 2010, IEEE Transactions on Computational Intelligence and AI in Games.

[25]  Kenneth R. Koedinger,et al.  Is Difficulty Overrated?: The Effects of Choice, Novelty and Suspense on Intrinsic Motivation in Educational Games , 2017, CHI.

[26]  Jesse Schell,et al.  The Art of Game Design: A book of lenses , 2019 .

[27]  Klaus Opwis,et al.  A Good Reason to Die: How Avatar Death and High Challenges Enable Positive Experiences , 2017, CHI.

[28]  Julian Togelius,et al.  Deep Learning for Video Game Playing , 2017, IEEE Transactions on Games.

[29]  Alex R. Wade,et al.  Modelling Early User-Game Interactions for Joint Estimation of Survival Time and Churn Probability , 2019, 2019 IEEE Conference on Games (CoG).

[30]  Mohsen Sardari,et al.  Winning Isn't Everything: Training Human-Like Agents for Playtesting and Game AI , 2019, ArXiv.

[31]  Anna Guitart,et al.  Churn Prediction in Mobile Social Games: Towards a Complete Assessment Using Survival Ensembles , 2016, 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA).

[32]  Julian Togelius,et al.  Automated Playtesting With Procedural Personas Through MCTS With Evolved Heuristics , 2018, IEEE Transactions on Games.

[33]  Ben Kirman,et al.  Learning curves: analysing pace and challenge in four successful puzzle games , 2014, CHI PLAY.

[34]  Samuel J. Gershman,et al.  Computational rationality: A converging paradigm for intelligence in brains, minds, and machines , 2015, Science.

[35]  Stefan Freyr Gudmundsson,et al.  Human-Like Playtesting with Deep Learning , 2018, 2018 IEEE Conference on Computational Intelligence and Games (CIG).

[36]  Wonjong Rhee,et al.  Churn prediction of mobile and online casual games using play log data , 2017, PloS one.

[37]  Erik Ragnar Poromaa Crushing Candy Crush : Predicting Human Success Rate in a Mobile Game using Monte-Carlo Tree Search , 2017 .

[38]  Paul J. Silvia,et al.  Curiosity and Interest: The Benefits of Thriving on Novelty and Challenge , 2009 .

[39]  Anna Guitart,et al.  Games and big data: A scalable multi-dimensional churn prediction model , 2017, 2017 IEEE Conference on Computational Intelligence and Games (CIG).

[40]  Kyung-Joong Kim,et al.  Game Data Mining Competition on Churn Prediction and Survival Analysis Using Commercial Game Log Data , 2018, IEEE Transactions on Games.

[41]  Andrew Nealen,et al.  Exploring Game Space Using Survival Analysis , 2015, FDG.

[42]  Marwan Mattar,et al.  Unity: A General Platform for Intelligent Agents , 2018, ArXiv.

[43]  Perttu Hämäläinen,et al.  Predicting Mid-Air Interaction Movements and Fatigue Using Deep Reinforcement Learning , 2020, CHI.

[44]  Paolo Burelli,et al.  Combining Sequential and Aggregated Data for Churn Prediction in Casual Freemium Games , 2019, 2019 IEEE Conference on Games (CoG).

[45]  Jari Takatalo,et al.  Review of Intrinsic Motivation in Simulation-based Game Testing , 2018, CHI.

[46]  L D NeysJoyce,et al.  Exploring persistence in gaming , 2014 .

[47]  Alec Radford,et al.  Proximal Policy Optimization Algorithms , 2017, ArXiv.

[48]  Aysu Betin Can,et al.  Enhancing the Monte Carlo Tree Search Algorithm for Video Game Testing , 2020, 2020 IEEE Conference on Games (CoG).

[49]  Shivakant Mishra,et al.  Mining Player In-game Time Spending Regularity for Churn Prediction in Free Online Games , 2019, 2019 IEEE Conference on Games (CoG).