The 2009 Mario AI Competition

Imitating human playing styles in Super Mario Bros

“…When Garry Kasparov lost to the IBM software/hardware Deep Blue, he famously complained that the computer played in an implausibly human-like manner, given that all other chess computers played in a distinctly machine-like fashion [3]. A similar tendency is true for the game that is used as a testbed in this article, Infinite Mario Bros: the AI that won the first annual Mario AI Competition played some levels in the game extremely well, but it's playing style looked so unnatural that a video of the winning controller playing a level became an Internet phenomenon [4]. Even when researchers set out specifically to create human-like controllers rather than well-playing ones, success is far from guaranteed.…”

“…The game has the advantage of being well known among the general public, This game was made into a benchmark ("The Mario AI Benchmark") for the Mario AI Championship 2 , a series of competitions that have been running in association with several international academic conferences on games and AI since 2009. The Mario AI Championship has four tracks: the Gameplay track, where competitors submit controllers that are judged on their capability to play the game as well as possible [4,17]; the Learning track, where submitted controllers are allowed to play each level 10000 times before being evaluated, in order to test the capability of the controllers to learn to play particular levels [17]; the Level Generation track, where competitors submit level generators that are judged on their capacity to generate engaging levels for human players [18]; and the Turing Test track, a recent addition where submitted controllers compete for being the most human-like, as judged by human spectators [2].…”

Learning Levels of Mario AI Using Genetic Algorithms

“…For instance, the paper published by Togelius et al summarizing the main results from the 2009 edition in the GamePlay track [10] describes the winner solution involving the use of the A* graph search algorithm, and briefly introduces other solutions using rule-based controllers, reactive controllers or finite state machines. Even when this paper referred to the Gameplay track, some solutions used learning algorithms such as genetic programming, stack-based virtual machines, and imitation or reinforcement learning; in some cases controllers are evolved using genetic algorithms.…”

“…This game implied a technical improvement in graphics, audio and gameplay over the original sidescroller, and introduced new characters like Yoshi. In 2009, the Mario AI Championship was introduced [10], aiming at developing intelligent agents able to complete levels of increasing difficulty of a game based on Infinite Mario Bros., a game based at the same time on Super Mario World (but with pseudo-randomly generated levels). In 2010, the Mario AI Championship introduced a new track: the Learning track, where an agent was intended to learn the best strategy to obtain the maximum score in a fixed level of the game, being able to play a maximum of 10,000 games of that same level before the competition in order to learn it.…”

Enhancing Level Difficulty and Additional Content in Platform Videogames through Graph Analysis

“… Infinite Mario Bros., an open-source platformer inspired by the classic videogame Super Mario Bros., which has been used frequently in academia, namely in the development of intelligent agents to control the main character [14] and automatic level generation [9].  The original version of the videogame Prince of Persia, which has unofficial level editors and technical details available online, as well as the recently released source code for the original Apple version of the game.…”