Tom Schaul

The 2014 General Video Game Playing Competition

This paper presents the framework, rules, games, controllers, and results of the first General Video Game Playing Competition, held at the IEEE Conference on Computational Intelligence and Games in 2014. The competition proposes the challenge of creating controllers for general video game play, where a single agent must be able to play many different games, some of them unknown to the participants at the time of submitting their entries. This test can be seen as an approximation of general artificial intelligence, as the amount of game-dependent heuristics needs to be severely limited. The games employed are stochastic real-time scenarios (where the time budget to provide the next action is measured in milliseconds) with different winning conditions, scoring mechanisms, sprite types, and available actions for the player. It is a responsibility of the agents to discover the mechanics of each game, the requirements to obtain a high score and the requisites to finally achieve victory. This paper describes all controllers submitted to the competition, with an in-depth description of four of them by their authors, including the winner and the runner-up entries of the contest. The paper also analyzes the performance of the different approaches submitted, and finally proposes future tracks for the competition.

An Extensible Description Language for Video Games

In this short paper, we propose a powerful new tool for conducting research on computational intelligence and games. “PyVGDL” is a simple, high-level, extensible description language for 2-D video games. It is based on defining locations and dynamics for simple building blocks (objects), together with local interaction effects. A rich ontology defines various controllers, object behaviors, passive effects (physics), and collision effects. It can be used to quickly design games, without having to deal with control structures. We show how the dynamics of many classical games can be generated from a few lines of PyVGDL. Furthermore, the accompanying software library permits parsing and instantly playing those games, visualized from a birds-eye or first-person viewpoint, and using them as benchmarks for learning algorithms.

Analyzing the robustness of general video game playing agents

This paper presents a study on the robustness and variability of performance of general video game-playing agents. Agents analyzed includes those that won the different legs of the 2014 and 2015 General Video Game AI Competitions, and two sample agents distributed with its framework. Initially, these agents are run in four games and ranked according to the rules of the competition. Then, different modifications to the reward signal of the games are proposed and noise is introduced in either the actions executed by the controller, their forward model, or both. Results show that it is possible to produce a significant change in the rankings by introducing the modifications proposed here. This is an important result because it enables the set of human-authored games to be automatically expanded by adding parameter-varied versions that add information and insight into the relative strengths of the agents under test. Results also show that some controllers perform well under almost all conditions, a testament to the robustness of the GVGAI benchmark.