October 22, 2015
Ph.D Candidate Aaron Isaken will be giving talks on AI and game design.
1) University of British Columbia on Oct 30th
Exploring Game Space Using Survival Analysis
Short Description of the Talk:
Game designers adjust game parameters to create an optimal experience for players — we call this high-dimensional set of unique game variants game space. To help designers explore game space and better understand the relationship between design and player experience, we present several methods that use survival analysis, a statistical technique normally used to model mechanical part failure, to find unique versions of varying difficulty. First, we focus on a parameterized version of Flappy Bird, a popular minimal score-based action game, and predict each variant’s difficulty using automatic play testing, Monte-Carlo simulation, a player model based on human motor skills (precision, reaction time, and actions per second), and exponential survival analysis of score histograms. Secondly, we use genetic algorithms and computational creativity to find the most unique variants in the Flappy Bird game space, discovering unique games that have a very different game feel merely by adjusting design parameters. Finally, we present methods for analyzing dynamic games that change their parameters over time, and demonstrate how skill learning by players quantitatively influences the effective difficulty of a game.
2) Full Indie Summit on Oct 31st
Designing Better Games Using Artificial Intelligence, Automated Play Testing, and Computational Creativity
Short Description of the Talk:
Making great games requires searching a massive space of all possible games to discover the hidden gems — which takes a huge amount of effort from designers and many hours of play testing. At the NYU Game Innovation Lab, we have developed methods to automate this process and assist designers in improving their games. I will show how we use artificial intelligence and player simulation to estimate the difficulty of different variants of Flappy Bird, using automated play testing to help a designer understand the effects of changing each game parameter. I also will show how we use genetic algorithms and computational creativity to find new variants of Flappy Bird that are as different from the original as possible. This talk will focus on practical methods coming from academic game labs which indie game designers can try to integrate into their own design process.
Aaron Isaksen is a PhD Candidate at the NYU Polytechnic School of Engineering Game Innovation Lab, studying automated game tuning, player modeling using probability distributions and survival analysis, and data-driven computer-aided game design. He has been working in the games industry since 2003, and is a graduate of UC Berkeley and MIT, where he worked in the Computer Graphics Group under Leonard McMillan and Steven Gortler on light field image-based rendering. Aaron is a founding partner of Indie Fund (www.indie-fund.com
), which aims to support the growth of games as a medium by helping indie developers get and stay financially independent, has co-created with his partnership AppAbove Games (www.appabovegames.com
) more than 10 indie mobile games, is an Advisory Board member for Fig, a crowdfunding platform, and is the Chairman of IndieBox (www.theindiebox.com
) which offers a subscription service for physical collectors editions of digital independent games. Aaron is also an organizer for IndieCade and an advisor for the NYU Game Center Incubator.