1 / 45

Measuring Entertainment and Automatic Generation of Entertaining Games

Measuring Entertainment and Automatic Generation of Entertaining Games. PhD Thesis Defense Zahid Halim Date: 23 rd November 2010 http:// ming.org.pk/zahid.htm . Supervised By: Dr. Rauf Baig. Presentation Outline. Introduction Problem Statement Thesis Objective Contribution

yahto
Télécharger la présentation

Measuring Entertainment and Automatic Generation of Entertaining Games

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Measuring Entertainment and Automatic Generation of Entertaining Games PhD Thesis Defense Zahid Halim Date: 23rd November 2010 http://ming.org.pk/zahid.htm Supervised By: Dr. Rauf Baig

  2. Presentation Outline • Introduction • Problem Statement • Thesis Objective • Contribution • Proposed Metrics • Board Based Games • Predator/prey Games • Conclusion • Future Plans • Major Achievements • Questions • Bibliography

  3. Problem Statement • Abundance of Games • Game Development Process • Issues • Quantifying entertainment • Writing new games/versions

  4. Thesis Objective • Define entertainment in games • Develop a quantitative measure of entertainment • Computational Intelligence to generate entertaining games • Verify evolved game’s entertainment

  5. Entertainment MetricsBoard Based Games • Duration of the Game • Intelligence for Playing the Game • Dynamism Exhibited by the Pieces • Usability of the Play Area

  6. Duration of the Game Metrics • Calculated by playing the game n times • Taking average number of moves over these n games • Maximum moves are fixed at 100

  7. Intelligence for Playing the Game Metrics • Number of wins of an intelligent controller over one making random moves • Higher number of wins against the random controller means that the game requires intelligence to be played and does not have too many frustrating dead ends • IK is 1 if intelligent controller wins the game otherwise it is 0

  8. Dynamism Exhibited by the Pieces Metrics • Game whose rules encourage greater dynamism of movement in its pieces would be more entertaining

  9. Usability of the Play Area Metrics • It is interesting to have the play area maximally utilized during the game

  10. Combined fitness • All chromosomes evaluated separately according to each of the four metrics • Then the population is sorted on each of the metrics separately • A rank based fitness is assigned to each chromosome. • The best chromosome assigned the highest fitness • Ranks multiplied by weights

  11. Entertainment Metrics Predator/prey Games • Duration of the Game • Appropriate Level of Challenge • Diversity • Usability

  12. Duration of the Game Metrics • In order to evolve games of short to medium duration we have fixed the upper bound of steps to 100 • 3 to 5 minute game if played with arrow keys • Premature death of agent possible • The death possibility of the agent should not be very high • Case the resulting games short and frustrating • Depend upon the agent playing the game

  13. Appropriate Level of ChallengeMetrics • High score, achieved easily and similarly too low • Not challenging enough • Game rules should provide an appropriate level of challenge • Factor of uncertainty in the rules of the game

  14. Diversity Metrics • The diversity of the game is based upon the diversity of the pieces in the game • The behavior of the moving pieces of the game should be sufficiently diverse so that it cannot be easily predicted

  15. Usability Metrics • It is interesting to have the play area maximally utilized during the game • If most of the moving pieces remain in a certain region of the play area then the resulting game may seem strange

  16. Combined fitness • All chromosomes evaluated separately according to each of the four metrics • Then the population is sorted on each of the metrics separately • A rank based fitness is assigned to each chromosome • The best chromosome assigned the highest fitness • Ranks multiplied by weights

  17. Board Based GamesSearch Space

  18. Chromosome Encoding

  19. Random Controller • Input: Game Board current state • Generate all legal moves • Store the moves in a queue • Shuffle the queue • If not mandatory to kill • Randomly select a move from the queue. • Else • Select a move that captures an opponent's piece, if such move exists • Otherwise, randomly select a move from the queue. • Output: Next move to take

  20. Min-Max based Controller • Input: Game Board current state • For each piece • priority=0 • For each piece • if is piece of honor • priority = priority +1 000 • if movement logic all directions • priority = priority + 8 • if movement logic diagonal Forward and Backward • priority = priority + 7 • if movement logic Straight Forward and Backward • priority = priority + 7 • if movement logic diagonal Forward • priority = priority + 6 • if movement logic Straight Forward • priority = priority + 6 • if movement logic L shaped • priority = priority + 5 • if capturing logic step into • priority = priority + 4 • if capturing logic step over • priority = priority + 3 • Count the number of pieces of Player A • Multiply the number of pieces of a type with its relevant priority • Count the number of pieces of Player B • Multiply the number of pieces of a type with its relevant priority • Calculate boardValue = WeightSumofA-WeightSumofB • Check if the Piece of Honour is dead add -1000 to boardValue • Check if the Piece of Honour is NOT dead add +1000 to boardValue • Output: boardValue

  21. Experimentation Setup • 1+1 Evolutionary Strategy (ES) • 10 chromosomes are randomly initialized • The evolutionary algorithm is run for 100 iterations • Mutation only with probability of 30 percent • One parent produce one child • Fitness difference is calculated • If it is greater than 4 (at least half times better) child is promoted to the next population

  22. Metrics values of one family

  23. Game Rules/Pieces Positions

  24. Learnability of Evolved Games 1/2 • Schmidhuber’s theory of artificial curiosity • Chellapilla’s architecture of the controller for checkers player • 5 layers in the ANN • Input with 64 neurons • First hidden layer with 91 neurons • Second hidden layer with 40 neurons • Third with 10 neurons • Output layer with 1 neuron. • Hyperbolic tangent function is used in each neuron • Connection weights range is [-2, 2]

  25. Learnability of Evolved Games 2/2 • The training of the ANN is done using co-evolution • GA population is initialized representing weight • Each individual played against randomly selected 5 others • Mutation only

  26. User Survey • Human user survey on 10 subjects • Chosen such that they have at least some level of interest towards computer games

  27. Predator/prey GamesSearch Space • 14 X 14 grid excluding the boundary walls. • Couple of walls at fixed positions and of size 7 cells  • There is one player controlled by the human player. • There are N (0-20)other pieces of M (1,2 and 3) types • Maximum duration 100 game steps • Finish game • Agent dies • Maximum score is achieved • Maximum game steps utilized •  Movement logic • No movement • Clockwise • Counter clockwise • Random • Random direction •  Collision logic • no effect • random relocation to a new location on the grid • death •  Scoring logic • +1, -1, 0

  28. Chromosome Encoding

  29. Rule Based Controller • The controller notes the nearest piece (if any) in each of the four directions moves one step towards the nearest score increasing piece • If there are no score increasing piece, step according to priority list • Move in the empty direction • If more than one such directions move towards farthest • Move towards score neutral piece • Move towards score decreasing piece • Move towards death causing piece

  30. ∆xr Connection Edges Connection Edges Connection Edges ∆yr Nu ∆xg Nd ∆yg Nl ∆xb Nr ∆yb Neural Network Based Controller • Multi-layer fully feed forward • 6 neurons in the input layer • 5 neurons in the hidden layer • 4 output layer neurons • Sigmoid activation function • Edges weights -5 to +5

  31. Experimentation Setup • 10 chromosomes are randomly initialized by the GA • One offspring is created for each chromosome • Duplicating it • Mutating any one of its gene • Results in 20 chromosomes from which 10 best chosen • 100 generations

  32. Duration of game Appropriate level of challenge Diversity Usability Combined Fitness

  33. Controller Learning Ability

  34. User Survey • 10 subjects • Conducted in two different sets on different days • Rule based controller • ANN based controller • Each individual was given 6 games • Play 2 times

  35. Conclusion • Identified the entertainment factors •  Introduced entertainment metrics • Board based genre of games • Video games • Predator/prey •  Entertainment factors dependent on genre • Automatic generation of entertaining games • Verification • Learnability of Evolved Games • User survey

  36. Limitations • Appropriate for offline mode • Processor intensive • Multiple times

  37. Future Plans • Multi objective genetic algorithm • Model the behavior of a particular human player • Evolving content for games against his/her playing patterns • Physical activating games for medical science

  38. Measuring Entertainment and Automatic Generation of Entertaining Games Thank you for your patience Questions

  39. Bibliography 1/7 • J.Schmidhuber, ”Developmental robotics, optimal artificial curiosity, creativity, music, and the fine arts”, Connection Science, vol. 18, pp.173–187, 2006 • N. Esposito, “A Short and Simple Definition of What a Videogame Is”, in proceedings of Digital Games Research Association (DiGRA), Vancouver, Canada, 16-20 June, 2005 • J.Smed and H.Hakonen, "Towards a Definition of a Computer Game", Technical Report, Computer Games Research Group, Department of Information Technology, University of Turku, Finland, 2005 • S. Lucas,” Evolving a neural network location evaluator to play Ms. Pac-Man. In Graham Kendall and Simon Lucas, editors”, Proceedings of the IEEE Symposium on Computational Intelligence and Games, pages 203–210, Essex University, Colchester, UK, 4–6 April 2005 • E. Norlig and L. Sonenberg, ” An approach to evaluating human characteristics in agents”, In Gabriela Lindemann, Daniel Moldt, Mario Paolucci, and Bin Yu, editors, in proceedings of the International Workshop on Regulated Agent-Based Systems: Theories and Applications , Bologna, Italy, 16th July 2002 • K. O. Stanley, N. Kohl, R. Sherony, and R. Miikkulainen,” Neuroevolution of an automobile crash warning system”, in proceedings of the Genetic and Evolutionary Computation Conference , Washington DC, USA, 25-29 June, 2005 • D. A. Jirenhed, G. Hesslow, and T.Ziemke, “Exploring internal simulation of perception in mobile robots”, in proceedings of the Fourth European Workshop on Advanced Mobile Robots, Lund, Sweden, 19-21 September, 2001 • C. Schlieder, P. Kiefer, S. Matyas,” Geogames: Designing Location-Based Games from Classic Board Games”, IEEE Intelligent Systems 21(5): 40-46, 2006 • H. Lund, T. Klitbo and C. Jessen, "Playware Technology for Physically Activating Play", Artificial Life and Robotics Journal, 165-174, 2005 • L. Davis, “Hybridization and Numerical Representation”, in L Davis (ed), The Handbook of Genetic Algorithms, Van Nostrand Reinhold, 1991, pp 61–71

  40. Bibliography 2/7 • I.Witten, and E.Frank, “Data Mining: Practical Machine Learning Tools and Techniques”, 2nd Edition, Morgan Kaufman, San Francisco, CA, 2004 • I. Rechenberg, “Evolutionsstrategie: OptimierungtechnischerSystemenachPrinzipienderBiologischen Evolution”, Frammann-HolzboogVerlag, Stuttgart, 1973 • K. Chellapilla, D. B. Fogel, ”Evolving an expert checkers playing program without using human expertise”, IEEE Trans. Evolutionary Computation 5(4): 422-428, 2001 • P. J. Angeline, G. M. Sauders, and J. B. Pollack, “ An evolutionary algorithm that constructs recurrent neural networks”, IEEE Transaction on Neural Networks, 5:54–65, 1994 • T. Back and H. P. Schwefel, “An overview of evolutionary algorithms for paramater optimization”, Evolutionary Computation, 1(1):1–23, 1993 • H. Iida, N. Takeshita, and J. Yoshimura, “A Metric for Entertainment of Board Games: Its Application for Evolution of Chess Variants”, In Nakatsu, R., and Hoshino, J., eds., Entertainment Computing: Technologies and Applications (Proceedings of IWEC 2002), pages 65-72. Boston, MA: Kluwer Academic Publishers, 2003 • A. Cincotti and H. Iida, “Outcome Uncertainty And Interestedness In Game-Playing: A Case Study Using Synchronized Hex”, In New Mathematics and Natural Computation (NMNC), vol. 02, issue 02, pages 173-181, 2006 • S. Retalis, “Creating Adaptive e-Learning Board Games for School Settings Using the ELG Environment”, Journal of Universal Computer Science: 2897-2908, 2008 • J. Togelius, R. D. Nardi, and S. M. Lucas,“Towards automatic personalised content creation for racing games”, in Proceedings of the IEEE Symposium on Computational Intelligence and Games, Piscataway, NJ, 1-5 April, 2007 • S. Sevenster, A.P. Engelbrecht, “GARTNet: A Genetic Algorithm for Routing in Telecommunications Networks”, Proceedings of CESA96 IMACS Multiconference on Computational Engineering in Systems Applications, Symposium on Control, Optimization and Supervision, Lille, Prance, Vol 2, 1996

  41. Bibliography 3/7 • G. N. Yannakakis , J. Hallam, ”Towards Optimizing Entertainment In Computer Games”, Applied Artificial Intelligence”, v.21 n.10, p.933-971, November 2007 • M. Gallagher and A. Ryan, “Learning to Play Pac-Man: An Evolutionary Rule-based Approach,” in proceedings of IEEE Congress on Evolutionary Computation, Canberra, Australia, 8-12 December. 2003 • G. Lankveld, P. Spronck, M. Rauterberg, “Difficulty Scaling through Incongruity”, in proceedings of the Fourth Artificial Intelligence and Interactive Digital Entertainment Conference, Stanford, California, 22–24 October, 2008 • J. Togelius and J. Schmidhuber, "An Experiment in Automatic Game Design", in Proceedings of IEEE Computational Intelligence and Games, Perth, Australia, 15 - 18 December, 2008 • J. Togelius, S. Karakovskiy, J. Koutnik and J. Schmidhuber, "Super Mario Evolution", in proceedings of IEEE Symposium on Computational Intelligence and Games, Milano, Italy, 7-10 September, 2009 • N. Lazzaro, “Why We Play Games: Four Keys to More Emotion Without Story”, in proceedings of Game Developers Conference, San Jose, California, 8th March, 2004 • C. Pedersen, J. Togelius, and G. N. Yannakakis, "Optimization of platform game levels for player experience,'' in proceedings of Artificial Intelligence and Interactive Digital Entertainment , Stanford, California, 14-16 October 2009 • K.Compton and M. Mateas, “Procedural Level Design for Platform Games”, in proceedings of 2nd Artificial Intelligence and Interactive Digital Entertainment Conference, Stanford, 20-23 June, California, 2006 • G. N. Yannakakis and J. Hallam, “Evolving Opponents for Interesting Interactive Computer Games”, in proceedings of the 8th International Conference on the Simulation of Adaptive Behavior,Los Angeles, USA, 13-17 July 2004 • B. D. Bryant and R. Miikkulainen, “Acquiring visibly intelligent behavior with example-guided neuroevolution”, in proceedings of the 22nd national conference on Artificial intelligence, Vancouver, British Columbia, Canada, 22-26 July, 2007

  42. Bibliography 4/7 • G. N. Yannakakis, and J. Hallam, "Entertainment Modeling through Physiology in Physical Play," International Journal of Human-Computer Studies, vol. 66, issue 10, pp. 741-755, October, 2008 • R. J. Orr and G. D. Abowd, “The smart floor: a mechanism for natural user identification and tracking”, in proceedings of Conference on Human Factors in Computing Systems, Netherlands, 1-6 April, 2000 • B. Richardson, K. Leydon, M. Fernstrom, and J. A. Paradiso, “Z-tiles: Building blocks for modular, pressure-sensing floorspaces”, in proceedings of Conference on Human Factors in Computing Systems, Vienna, Austria, 24-29 April, 2004 • H. H. Lund, T. Klitbo, and C. Jessen, “Playware technology for physically activating play”, Artifical Life and Robotics Journal, vol. 9, no. 4, pp. 165–174, 2005 • K.Grønbæk, O.S.Iversen, K.J.Kortbek, K.R.Nielsen, and L. Aagaard, " iGameFloor - a Platform for Co-Located Collaborative Games", in proceedings of the International Conference on Advances in Computer Entertainment, Salzburg, Austria, 13.-15. June, 2007 • N. Beume, H. Danielsiek, C. Eichhorn, B. Naujoks, M. Preuss, K. Stiller, S. Wessing, “Measuring Flow as Concept for Detecting Game Fun in the Pac-Man Game”, in proceedings of the IEEE Congress on Evolutionary Computation, CEC 2008, Hong Kong, China, 1-6 June, 2008 • K.O.Stanley, B.D.Bryant, R.Miikkulainen, “Evolving neural network agents in the nero video game”, in proceedings of IEEE Symposium on Computational Intelligence and Games, Milan, September, 2009 • B. D. Bryant and R. Miikkulainen, "Evolving stochastic controller networks for intelligent game agents," in proceedings of IEEE Congress on Evolutionary Computation, San Francisco, California, 26-29 June, 2006 • B. D. Bryant and R. Miikkulainen, “Evolving visibly intelligent behavior for embedded game agents”, PhD thesis, Department of Computer Sciences, University of Texas at Austin, Austin, TX, 2006 • D.L. Roberts, C.R. Strong and C.LIsbell,” Estimating player satisfaction through the author's eyes”, in proceedings of the Second Workshop on Optimizing Player Satisfaction, California, USA,6-8 July, 2007

  43. Bibliography 5/7 • B. D. Bryant and R. Miikkulainen, ”Exploiting sensor symmetries in example-based training for intelligent agents”, In Louis, S. J., and Kendall, G., eds., Proceeedings of the IEEE Symposium on Computational Intelligence and Games, Reno Tahoe, 22-24 May, 2006 • B. D. Bryant and R. Miikkulainen, “Neuroevolution for adaptive teams,” in proceedings of IEEE International Conference on Electronic Commerce, Newport Beach, CA, USA, 24-27 June 2003 • N. Beume, T. Hein, B. Naujoks, N. Piatkowski, M. Preuss, M. and S. Wessing, “Intelligent Anti-Grouping in Real-Time Strategy Games”, in proceedings of IEEE Symposium on Computational Intelligence and Games Perth, Australia, 15 - 18 December, 2008 • M. Csíkszentmihályi, “Flow: The Psychology of Optimal Experience”, Harper and Row. ISBN 0-06-092043-2, 1991. • A. Bezek, “ModelingMultiagent Games Using Action Graphs”, in proceedings of Modeling Other Agents from Observations, New York, USA, 19th July, 2004 • B. Christian Bauckhage, C. Gorman, Thurau & M. Humphrys, "Learning Human Behavior from Analyzing Activities in Virtual Environments", MMI interaktiv journal, Issue: 12 Pages: 3-17, 2007 • T. Henderson and S. Bhatti, “Modelling user behaviour in networked games”, Multimedia 2001, June 2001 • H. Barber and D. Kudenko, “A User Model for the Generation of Dilemma-based Interactive Narratives”, in proceeding of third Artificial Intelligence for Interactive Digital Entertainment Conference, California, USA, 6-8 June, 2007 • R. Hunicke and V. Chapman, “ AI for Dynamic Difficulty Adjustment in Games”, in proceedings of the Challenges in Game AI Workshop, Nineteenth National Conference on Artificial Intelligence, San Jose, California, USA, 25-29 July, 2004 • N.A.Taatgen, and J.R.Anderson, “Why do children learn to say “Broke”? A model of learning the past tense without feedback.”, Cognition, 86(2),123-155, December 2002

  44. Bibliography 6/7 • D. Thue, V. Bulitko, M. Spetch, E. Wasylishen,” Interactive Storytelling: A Player Modelling Approach”, in proceeding of third Artificial Intelligence for Interactive Digital Entertainment Conference, California, USA, 6-8 June, 2007 • M. Kriegel, R. Aylett, J. Dias and A. Paiv, ”Learning Player Preferences to Inform Delayed Authoring”, in AAAI Symposium on. Intelligent Narrative Technologies. Westin Arlington Gateway, Arlington, Virginia, 9-11 November, 2007 • P. Rani, N. Sarkar and C. Liu, “Maintaining Optimal Challenge in Computer Games Through Real-Time Physiological Feedback”, 11th Human Computer Interaction International, Vegas, Nevada, USA, 22-27 July 2005 • S. W. McQuiggan, S. Lee and J. C. Lester, “Predicting User Physiological Response for Interactive Environments: An Inductive Approach”, in proceeding of third Artificial Intelligence for Interactive Digital Entertainment Conference, California, USA, 20-23 June, 2007 • R. Picard, E. Vyzas and J. Healey, ”Toward machine emotional intelligence: analysis of affective physiological state”, in IEEE Transactions Pattern Analysis and Machine Intelligence 23(10): 1185-1191,2001 • K. P. Pomsta and H. Pain, “Providing Cognitive and Affective Scaffolding through Teaching Strategies”, in proceedings of 7th International Conference on Intelligent Tutoring Systems, Maceió, Brazil, 31st August, 2004 • H.Prendinger, S.Mayer, J.Mori, and M.Ishizuka, “Persona effect revisited: Using bio-signals to measure and reflect the impact of character-based interfaces”, in proceedings of 4th International Conference on Intelligent Virtual Agents, KlosterIrsee, Germany, 15-17 September , 2003 • N. A. Taatgen, M. V. Oploo, J. Braaksma, J. Niemantsverdriet,” How to Construct a Believable Opponent using Cognitive Modeling in the Game of Set”, in proceedings of the fifth international conference on cognitive modelling, Bamberg, Germany. 9-12 April, 2003 • N. Beume, B. Naujoks, G. Neugebauer, N. Piatkowski and M. Preuss , “CI enlivened ghosts for Pac-Man”, in proceedings of IEEE Computational Intelligence And Games, Perth, Australia, 15-18 September, 2008

  45. Bibliography 7/7 • G. V. Lankveld, P. Spronck, and M. Rauterberg, ”Incongruity-Based Adaptive Game Balancing”, Advances In Computer Games, Pamplona, Spain, 11 - 13 May 13, 2009 • N. Beume, T. Hein, B. Naujoks, N. Piatkowski, M. Preuss, S. Wessing, “Intelligent Anti-Grouping in Real-Time Strategy Games”, In Proceedings of IEEE Computational Intelligence and Games, Perth, Australia, 15 - 18 December, 2008 • H. Danielsiek, R. Stüer, A. Thom, N. Beume, B. Naujoks and M. Preuss, “Intelligent Moving of Groups in Real-Time Strategy Games”, in proceedings of IEEE Computational Intelligence And Games, Perth, Australia, 15-18 September, 2008 • M. Csikszentmihalyi and I. Csikszentmihalyi, “Introduction to Part IV in Optimal Experience”, Psychological Studies of Flow in Consciousness, Cambridge, UK: Cambridge University Press. 1988 • T. W. Malone, “ What makes computer games fun?”, ACM SIGSOC Bulletin, Volume 13 , Issue 2-3 1982 • Presentation Chess Pieces Image, https://swansea-edunet.gov.uk/en/schools/Pontarddulais/PageImages/Chess%20Piece.jpg, 05-11-2010 • Presentation background Images, http://www.presentermedia.com, 05-11-2010 • R. Koster, “ A Theory of Fun for Game Design”, Paraglyph Press, 2005 • D. H. Ackley and M. L. Littman,”Interactions between learning and evolution”, In C. G. Langton, C. Taylor, J. D. Farmer, and S. Rasmussen, editors, Artificial Life II, pages 478–507, Reading, Sante Fe Institute Studies in the Sciences and Complexity, Addison-Wesley, MA, 1992

More Related