Cristina Guerrero-Romero

General​ ​video​ ​game​ ​playing​ ​agents​ ​to​ ​evaluate​ ​automatic​ ​generated​ ​levels

Procedural Content generation (PCG) refers to the generation of game content via automated processes. The validity of these artifacts should be checked to ensure their playability. Common approaches are game-specific and, although they ensure the playability of the generated content, they are limited by the game they are defined for. To solve this limitation, a strong connection between PCG and General Video Game (referring to algorithms that do not take game-specific knowledge into account) could be established, as general controllers could be used to validate content generated for any game.

The ultimate goal of Cris’ research is creating a system capable of using General Video Game (GVG) agents to play and evaluate new automated generated levels for games. The evaluator would be formed by different controllers and would be capable of analyzing a provided level of a game without having any previous knowledge about the level or the game. The idea is giving an answer that would determine if a level should be included into the game or not, based on the expectations.

Cris is a Spaniard in love with London, where she’s been living since 2013. She studied a BE in Computer Engineering at Universidad Autónoma de Madrid (Spain) and worked as web developer for a couple of years before taking the decision of changing to the exciting world of AI and games. Based in Queen Mary University of London, her research interests include General Video Game Playing (GVGP), Procedural Content Generation (PCG) and automatic evaluation. She tries to keep active outside the sedentary PhD live, mostly walking around the city, cycling and attempting different sports. Random facts are her fascination with swords and that her chosen superpower would be teleportation.