Post Doctoral Researcher

I am a post doctoral researcher in the College of Arts, Media, and Design at Northeastern University under the advisement of Dr. Casper Harteveld. My thesis work was on using Markov models for procedural content generation (PCG), and more generally the application of machine learning to PCG. In my current role I am exploring the gamification of behavioral and psychological surveys. Additionally, I am exploring the use of those surveys and other techniques for player and user modeling in the context of StudyCrafter, an platform for creating and conducting research projects. After I complete my post doc, I will be looking for professorship positions in the the fields of Artificial Intelligence, Games, and Machine Learning.

Journal Publications 

Summerville, Adam, Sam Snodgrass, Matthew Guzdial, Christoffer Holmgård, Amy K. Hoover, Aaron Isaksen, Andy Nealen, and Julian Togelius. Procedural Content Generation via Machine Learning (PCGML). arXiv preprint arXiv:1702.00539. 2017. (submitted to TCIAIG) [PDF]

Snodgrass, Sam, and Santiago Ontanón. Learning to Generate Video Game Maps Using Markov Models. IEEE Transactions on Computational Intelligence and AI in Games. 2016. [PDF]

Conference Papers/Technical Reports

Snodgrass, Sam, Adam Summerville, and Santiago Ontanón. Studying the Effects of Training Data on Machine Learning-based Procedural Content Generation. Thirteenth Artificial Intelligence and Interactive Digital Entertainment Conference. 2017. [PDF]

Snodgrass, Sam, Santiago Ontanón. Leveraging Multi-Layer Level Representations for Puzzle-Platformer Level Generation. Fifth Experimental AI in Games (EXAG) workshop. 2017. [PDF]

Summerville, Adam, Julian R. H. Marino, Sam Snodgrass, Santiago Ontanón, and Levi H. S. Lelis. Understanding Mario: An Evaluation of Design Metrics for Platformers. Twelfth International Conference on the Foundations of Digital Games. 2017. [PDF]

Snodgrass, Sam and Santiago Ontanón. Procedural Level Generation using Multi-layer Level Representations with MdMCs. Computational Intelligence and Games (CIG), 2017 IEEE Conference on. IEEE, 2017. [PDF]

Snodgrass, Sam and Santiago Ontanón. Player Movement Models for Video Game Level Generation. Twenty-Sixth International Joint Conference on Artificial Intelligence. 2017. [PDF]

Ontañón, Santiago, Yi-Ching Li, Sam Snodgrass, Flaura K. Winston, and Avelino J. Gonzalez Learning to Predict Driver Behavior from Observation Proceedings of the AAAI Spring Symposium: Learning from Observation of Humans, Stanford, USA. 2017. [PDF]

Snodgrass, Sam, and Santiago Ontanón. An Approach to Domain Transfer in Procedural Content Generation of Two-Dimensional Videogame Levels. Twelfth Artificial Intelligence and Interactive Digital Entertainment Conference. 2016. [PDF]

Snodgrass, Sam, and Santiago Ontanón. Controllable Procedural Content Generation via Constrained Multi-Dimensional Markov Chain Sampling. Twenty-Fifth International Joint Conference on Artificial Intelligence. 2016. [PDF]

Summerville, Adam James, Sam Snodgrass, Santiago Ontanón, and Michael MateasThe VGLC: The Video Game Level Corpus. Proceedings of the 7th Workshop on Procedural Content Generation at 1st Joint International Conference of DiGRA and FDG. 2016. [PDF]

Snodgrass, Sam, and Santiago Ontañón. A Hierarchical MdMC Approach to 2D Video Game Map Generation. Eleventh Artificial Intelligence and Interactive Digital Entertainment Conference. 2015. [PDF]

Snodgrass, Sam, and Santiago Ontañón. A Hierarchical Approach to Generating Maps Using Markov Chains. Tenth Artificial Intelligence and Interactive Digital Entertainment Conference. 2014. [PDF]

Snodgrass, Sam, Benjamin Goldberg, Ariel Evans, Brandon Packard, Cathy Lu, and Jichen Zhu. Extended abstract for Canvas Obscura. Proceedings of the first ACM SIGCHI annual symposium on Computer-human interaction in play. ACM, 2014. [PDF]

Snodgrass, Sam, and David W. Aha. System Model Formulation Using Markov Chains. (Technical Note AIC-14-170). Washington, DC: Naval Research Laboratory, Navy Center for Applied Research in AI. 2014.[Technical Report]

Ontañón, Santiago, Yi-Ching Li, Sam Snodgrass, Dana Bonfiglio, Flaura K. Winston, Catherine McDonald, and Avelino J. Gonzalez. Case-Based Prediction of Teen Driver Behavior and Skill. Case-Based Reasoning Research and Development. Springer International Publishing, 2014. 375-389. [PDF]

Snodgrass, Sam, and Santiago Ontañón. Experiments in Map Generation using Markov Chains. Ninth International Conference on the Foundations of Digital Games. 2014. [PDF]

Snodgrass, Sam, and Santiago Ontañón. Generating Maps Using Markov Chains. AIIDE Workshop on Artificial Intelligence and Game Aesthetics. 2013. [PDF]

Projects
  • MdMCs: Multi-dimensional Markov chains for the procedural generation of video game levels. 
    We have explored many extensions to this approach including
    • Domain transfer from one game to another
    • Constrained sampling approaches to ensure specific qualities in output levels
    • Multi-layer level representation to more fully capture level information
  • VGLC: The video game level corpus, a collection of video game levels meant to be used as
    training data for procedural content generation vial machine learning (PCGML) approaches. 
  • Modeling human driving behavior: An NSF funded project focused on learning models of 
    human drivers, and using those models to predict detrimental driving behaviors, as well as
    to predict different classes of drivers (e.g., experienced or novice).
  • Canvas Obscura: A horror game with procedurally generated level layouts.

Service and Community Involvement
  • Reviewer/Subreviewer
    • Knowledge Extraction from Games Workshop, 2017. (KEG 2017)
    • Transactions on Computational Intelligence and Artificial Intelligence in Games, 2017. (TCIAIG 2017)
    • Ninth Foundations of Digital Games, 2017. (FDG 2017)
    • Eighth Workshop on Procedural Content Generation, 2017. (PCG 2017)
    • Third Computational Creativity and Games Workshop, 2017. (CCGW 2017)
    • Eighth Foundations of Digital Games, 2016. (FDG 2016) 
    • Seventh Foundations of Digital Games, 2015. (FDG 2015)
  • Tutorials
    • Snodgrass, Sam and Adam Summerville. Tutorial on Procedural Content Generation via Machine Learning. Computational
      Intelligence in Games. 2017.
  • Mentoring
    • Research Experience for Teachers (REThink)
      • Guided two high school teachers through a Summer-long project on procedural content generation.
    • Mentored 3 high school students (on 3 separate occasions) as they worked on projects related to procedural content generation

Teaching Assistantships at Drexel University

 Term  Course 
 Spring 2014  CS 380: Artificial Intelligence
 Winter 2014  CS 260: Data Structures
 Fall 2013
 CS 380: Artificial Intelligence
 Spring 2013  CS 123: Computation Lab 3
 Winter 2013  CS 122: Computation Lab 2
 Fall 2012  CS 121: Computation Lab 1

Personal Links