Deep Static and Dynamic Level Analysis: A Study on Infinite Mario

Automatic analysis of game levels can provide assistance to game designers and procedural content generation. We introduce a static-dynamic scale to categorize level analysis strategies, which captures the extent that the analysis depends on player simulation. Due to its ability to automatically learn intermediate representations for the task, a convolutional neural network (CNN) provides a general tool for both types of analysis. In this paper, we explore the use of CNN to analyze 1,437 Infinite Mario levels. We further propose a deep reinforcement learning technique for dynamic analysis, which allows the simulated player to pay a penalty to reduce error in its control. We empirically demonstrate the effectiveness of our techniques and complementarity of dynamic and static analysis.

[1]  Lawrence D. Jackel,et al.  Backpropagation Applied to Handwritten Zip Code Recognition , 1989, Neural Computation.

[2]  Kevin J. Lang A time delay neural network architecture for speech recognition , 1989 .

[3]  Hiroyuki Iida,et al.  A metric for entertainment of boardgames: its implication for evolution of chess variants , 2002, IWEC.

[4]  Robert Zubek,et al.  MDA : A Formal Approach to Game Design and Game Research , 2004 .

[5]  Staffan Björk,et al.  Patterns In Game Design , 2004 .

[6]  Peta Wyeth,et al.  GameFlow: a model for evaluating player enjoyment in games , 2005, CIE.

[7]  Julian Togelius,et al.  Towards automatic personalised content creation for racing games , 2007, 2007 IEEE Symposium on Computational Intelligence and Games.

[8]  Julian Togelius,et al.  An experiment in automatic game design , 2008, 2008 IEEE Symposium On Computational Intelligence and Games.

[9]  M. Larzelere,et al.  Stress and health. , 2008, Primary care.

[10]  Julian Togelius,et al.  Modeling player experience in Super Mario Bros , 2009, 2009 IEEE Symposium on Computational Intelligence and Games.

[11]  Julian Togelius,et al.  Towards Automatic Personalized Content Generation for Platform Games , 2010, AIIDE.

[12]  Julian Togelius,et al.  The 2009 Mario AI Competition , 2010, IEEE Congress on Evolutionary Computation.

[13]  Michael Mateas,et al.  Tanagra: a mixed-initiative level design tool , 2010, FDG.

[14]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[15]  E. James Whitehead,et al.  Design patterns in FPS levels , 2010, FDG.

[16]  Simon Colton,et al.  Aesthetic Considerations for Automated Platformer Design , 2012, AIIDE.

[17]  Simon Colton,et al.  Initial Results from Co-operative Co-evolution for Automated Platformer Design , 2012, EvoApplications.

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

[19]  Julian Togelius,et al.  Towards a Generic Method of Evaluating Game Levels , 2013, AIIDE.

[20]  Zoran Popovic,et al.  Automated redesign of local playspace properties , 2013, FDG.

[21]  Roger Crawfis,et al.  Optimal cover placement against static enemy positions , 2013, FDG.

[22]  Julian Togelius,et al.  A multi-level level generator , 2014, 2014 IEEE Conference on Computational Intelligence and Games.

[23]  Clark Verbrugge,et al.  Measuring risk in stealth games , 2014, FDG.

[24]  Stefan Carlsson,et al.  CNN Features Off-the-Shelf: An Astounding Baseline for Recognition , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[25]  Julian Togelius,et al.  A comparative evaluation of procedural level generators in the Mario AI framework , 2014, FDG.

[26]  Petros Faloutsos,et al.  Characterizing and optimizing game level difficulty , 2014, MIG 2014.

[27]  Clark Verbrugge,et al.  An Algorithmic Approach to Decorative Content Placement , 2021, Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment.

[28]  Alessandro Canossa,et al.  Towards a Procedural Evaluation Technique: Metrics for Level Design , 2015, FDG.

[29]  Gillian Smith,et al.  Formalizing Non-Formalism: Breaking the Rules of Automated Game Design , 2015, FDG.

[30]  Levi Lelis,et al.  An Empirical Evaluation of Evaluation Metrics of Procedurally Generated Mario Levels , 2021, AIIDE.

[31]  Ya'akov Gal,et al.  Human computation for procedural content generation in platform games , 2015, 2015 IEEE Conference on Computational Intelligence and Games (CIG).

[32]  Michael Mateas,et al.  Super Mario as a String: Platformer Level Generation Via LSTMs , 2016, DiGRA/FDG.