Postdoctoral Researcher at Maastricht University

Member of the Department of Data Science and Knowledge Engineering


I am currently a Postdoctoral Researcher at Maastricht University's Department of Data Science and Knowledge Engineering (DKE), working on the Digital Ludeme Project.

Curriculum Vitae

Research Interests

My research interests centre predominantly around the use of Artificial Intelligence and Machine Learning techniques for games. This includes research that uses AI techniques to help create better games, but also how we can utilise games as a testbed for developing solutions to real-world problems.

I am currently working as a postdoctoral researcher on the Digital Ludeme Project, which aims to improve our understanding of traditional games using modern AI techniques. This project can be summarised as a computational study of the world's traditional strategy games throughout recorded human history, hoping to chart their historical development and explore their role in the development of human culture and the spread of mathematical ideas.

My PhD research focussed on implementing various AI techniques for physics-based video games and simulations. This involved developing AI agents that can reason and interact within a physical environment, as well as generating content that satisfies the physical limitations of such environments. During this time I worked on multiple research papers for the physics-based puzzle game Angry Birds, and was a key organiser for two AI competitions centred around this game (AIBIRDS).

I have also conducted research into many other fields of game AI, including topics such as general video game agents, procedural content generation, deep learning networks, hyper-agent development, computational complexity analysis, meta-learning, sketch-based generation, deceptive game design, symmetry detection and augmented reality.

Digital Ludeme Project

Project involving computational study of the world's traditional strategy games throughout recorded human history.


Competition Organiser:


Ludii AI Competition

Competition focused on developing general game playing agents for the Ludii system.

Angry Birds AI Competition

Competition focused on developing intelligent agents to solve Angry Birds levels.

Angry Birds Level Generation Competition

Competition involving procedurally generating levels for Angry Birds.

Publications

2021

M. Stephenson, D. Soemers, E. Piette, C. Browne, General Game Heuristic Prediction Based on Ludeme Descriptions, IEEE Conference on Games (IEEE-COG'21), Copenhagen, Denmark, August 2021. (pdf file)

E. Piette, M. Stephenson, D. Soemers, C. Browne, General Board Game Concepts, IEEE Conference on Games (IEEE-COG'21), Copenhagen, Denmark, August 2021. (pdf file)

C. Gamage, M. Stephenson, V. Pinto, J. Renz, Deceptive Level Generation for Angry Birds, IEEE Conference on Games (IEEE-COG'21), Copenhagen, Denmark, August 2021. (pdf file)

C. Gamage, V. Pinto, C. Xue, M. Stephenson, P. Zhang, J. Renz, Novelty Generation Framework for AI Agents in Angry Birds Style Physics Games, IEEE Conference on Games (IEEE-COG'21), Copenhagen, Denmark, August 2021. (pdf file)

D. Soemers, V. Mella, E. Piette, M. Stephenson, C. Browne, O. Teytaud, Transfer of Fully Convolutional Policy-Value Networks Between Games and Game Variants, CoRR arXiv:2102.12375, 2021. [bib] (pdf file)

W. Crist, M. Stephenson, The Digital Ludeme Project Games Database: Compiling Evidence to Reconstruct Historical Games, Board Game Studies (BGS'21), Paris, France, April 2021. [bib] (pdf file) (video presentation)

2020

D. Soemers, E. Piette, M. Stephenson, C. Browne, Manipulating the Distributions of Experience used for Self-Play Learning in Expert Iteration, IEEE Conference on Games (IEEE-COG'20), Osaka, Japan, August 2020. [bib] (pdf file) (video presentation)

E. Piette, D. Soemers, M. Stephenson, C. Sironi, M. Winands, C. Browne, Ludii - The Ludemic General Game System, European Conference on Artificial Intelligence (ECAI'20), Santiago de Compestela, Spain, August 2020. [bib] (pdf file) (video presentation)

M. Stephenson, J. Renz, X. Ge, The Computational Complexity of Angry Birds (Extended Abstract), International Joint Conference on Artificial Intelligence (IJCAI'20), Yokohama, Japan, July, 2020. [bib] (pdf file) (video presentation)

M. Stephenson, D. Anderson, A. Khalifa, J. Levine, J. Renz, J. Togelius, C. Salge, A Continuous Information Gain Measure to Find the Most Discriminatory Problems for AI Benchmarking, IEEE World Congress on Computational Intelligence (WCCI'20), IEEE Congress on Evolutionary Computation (CEC'20), Glasgow, United Kingdom, July, 2020. [bib] (pdf file) (video presentation)

M. Stephenson, J. Renz, X. Ge, The Computational Complexity of Angry Birds, Artificial Intelligence (AIJ), 2020. [bib] (pdf file) Published version of the paper can be accessed here.

2019

J. Renz, X. Ge, M. Stephenson, P. Zhang, AI meets Angry Birds, Nature Machine Intelligence, 2019. [bib] (pdf file)

P. Bontrager, A. Khalifa, D. Anderson, M. Stephenson, C. Salge, J. Togelius, “Superstition” in the Network: Deep Reinforcement Learning Plays Deceptive Games, The Fifteenth Annual AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE'19), Atlanta, GA, October 2019. [bib] (pdf file)

R. Gaina, M. Stephenson, “Did You Hear That?” Learning to Play Video Games from Audio Cues, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

M. Stephenson, E. Piette, D. Soemers, C. Browne, Ludii as a Competition Platform, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

C. Piette, E. Piette, M. Stephenson, D. Soemers, C. Browne, Ludii and XCSP: Playing and Solving Logic Puzzles, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

M. Stephenson, E. Piette, D. Soemers, C. Browne, An Overview of the Ludii General Game System, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

E. Piette, M. Stephenson, D. Soemers, C. Browne, An Empirical Evaluation of Two General Game Systems: Ludii and RBG, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

D. Soemers, E. Piette, M. Stephenson, C. Browne, Learning Policies from Self-Play with Policy Gradients and MCTS Value Estimates, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

T. Liu, J. Renz, P. Zhang, M. Stephenson, Using Restart Heuristics to Improve Agent Performance in Angry Birds, IEEE Conference on Games (IEEE-COG'19), London, United Kingdom, August 2019. [bib] (pdf file)

C. Browne, M. Stephenson, E. Piette, D. Soemers, A Practical Introduction to the Ludii General Game System, Proceedings of Advances in Computer Games (ACG'19), Macau, Springer, August 2019. [bib] (pdf file)

M. Stephenson, Generation and Analysis of Content for Physics-Based Video Games, PhD Thesis, July 2019. [bib] (pdf file)

E. Piette, D. Soemers, M. Stephenson, C. Sironi, M. Winands, C. Browne, Ludii - The Ludemic General Game System, Conférence Nationale en Intelligence Artificielle, Toulouse, France, July 2019. [bib] (English pdf file) (French pdf file)

M. Stephenson, E. Piette, C. Browne, Teaching and Learning with LUDII, Board Game Studies (BGS'19), Bologna, Italy, May 2019. [bib] (pdf file)

C. Browne, D. Soemers, E. Piette, M. Stephenson, M. Conrad, W. Crist, T. Depaulis, E. Duggan, F. Horn, S. Kelk, S. Lucas, J. Pedro Neto, D. Parlett, A. Saffidine, U. Schädler, J. Nuno Silva, A. de Voogt, M. Winands, Foundations of Digital Archæoludology, CoRR arXiv:1905.13516, 2019. [bib] (pdf file)

M. Stephenson, J. Renz, X. Ge, P. Zhang, Generating Stable, Building Block Structures from Sketches, IEEE Transactions on Games (TOG), 2019. [bib] (pdf file)

M. Stephenson, D. Perez-Liebana, M. Nelson, A. Khalifa, A. Zook, Game Complexity vs Strategic Depth, NII Shonan Meeting 130, Artificial General Intelligence in Games: Where Play Meets Design and User Experience, Shonan, Japan, March 2019. [bib] (pdf file)

M. Stephenson, J. Renz, Agent-Based Adaptive Level Generation for Dynamic Difficulty Adjustment in Angry Birds, Workshop on Games and Simulations for Artificial Intelligence at AAAI-2019, Honolulu, HI, January 2019. [bib] (pdf file)

2018

M. Stephenson, J. Renz, X. Ge, L. Ferreira, J. Togelius, P. Zhang, The 2017 AIBIRDS Level Generation Competition, IEEE Transactions on Games (TOG), 2018. [bib] (pdf file)

M. Stephenson, J. Renz, X. Ge, P. Zhang, The 2017 AIBIRDS Competition, CoRR arXiv:1803.05156, 2018. [bib] (pdf file)

M. Stephenson, J. Renz, Deceptive Angry Birds: Towards Smarter Game-Playing Agents, The Twelfth International Conference on the Foundations of Digital Games (FDG'18), Malmo, Sweden, August 2018, (honourable mention). [bib] (pdf file)

M. Stephenson, J. Renz, X. Ge, P. Zhang, Generating Stable, Building Block Structures from Sketches, Computer Games Workshop at IJCAI-ECAI'18, Stockholm, Sweden, July 2018. [bib] (pdf file)

D. Anderson, M. Stephenson, J. Togelius, C. Salge, J. Levine, J. Renz, Deceptive Games, Proceedings of EvoGames 2018 (EvoStar'18), Parma, Italy, April 2018. [bib] (pdf file)

2017

M. Stephenson, J. Renz, X. Ge, The Computational Complexity of Angry Birds and Similar Physics-Simulation Games, The Thirteenth Annual AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE'17), Snowbird, UT, October 2017. [bib] (pdf file)

M. Stephenson, J. Renz, Creating a Hyper-Agent for Solving Angry Birds Levels, The Thirteenth Annual AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE'17), Snowbird, UT, October 2017. [bib] (pdf file)

M. Stephenson, J. Renz, Generating Varied, Stable and Solvable Levels for Angry Birds Style Physics Games, IEEE Computational Intelligence and Games Conference 2017 (IEEE-CIG'17), New York, NY, August 2017. [bib] (pdf file)

D. Perez-Liebana, M. Stephenson, R. Gaina, J. Renz, S. M. Lucas, Introducing Real World Physics and Macro-Actions to General Video Game AI, IEEE Computational Intelligence and Games Conference 2017 (IEEE-CIG'17), New York, NY, August 2017. [bib] (pdf file)

2016

M. Stephenson, J. Renz, Procedural Generation of Levels for Angry Birds Style Physics Games, The Twelfth Annual AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE'16), Burlingame, CA, October 2016. [bib] (pdf file)

M. Stephenson, J. Renz, Procedural Generation of Complex Stable Structures for Angry Birds Levels, IEEE Computational Intelligence and Games Conference 2016 (IEEE-CIG'16), Santorini, Greece, September 2016. [bib] (pdf file)

2015

M. Stephenson, A. Clark, R. Green, Novel Methods for Reflective Symmetry Detection in Scanned 3D Models, The 30th International Conference on Image and Vision Computing New Zealand (IVCNZ'15), Auckland, New Zealand, November 2015. [bib] (pdf file)

M. Stephenson, A. Clark, R. Green, Novel Methods for Reflective Symmetry Detection in Scanned 3D Models (Honours Thesis), University of Canterbury, Christchurch, New Zealand, 2015. [bib] (pdf file)

Software

Iratus Aves (MSGv2.0)
Winning entry for the 2017 and 2018 Angry Birds level generation competitions.
https://github.com/stepmat/IratusAves

Angry Birds Sketch-based Generator
Generates structures for Science Birds (Angry Birds) from sketches of rectilinear polygons, used for research paper presented at CGW18 and published in Transactions on Games (TOG).
https://github.com/stepmat/ScienceBirds_sketch_generation

Deceptive Angry Birds levels
Set of 30 deceptive Angry Birds levels, used for research paper presented at FDG18.
https://github.com/stepmat/DeceptiveAngryBirds

Rectilinear Polygon Generator
Generates a random rectilinear polygon, formed by multiple overlapping rectangles.
https://github.com/stepmat/rectilinear_polygon_generator

Angry Birds level converter
Converts levels between the Science-Birds .xml format, and the official Angry Birds .json format.
https://github.com/stepmat/AIBIRDS_level_converter

Contact Me

Email:
matthew.stephenson@maastrichtuniversity.nl

Address:
Room 0.014
Tapijn Building Z
Department of Data Science and Knowledge Engineering
Maastricht University
Maastricht, 6211 KC
Limburg, Netherlands