Open Loop Search for General Video Game Playing

Abstract

General Video Game Playing is a sub-field of Game Artificial Intelligence, where the goal is to find algorithms capable of playing many different real-time games, some of them unknown a priori. In this scenario, the presence of domain knowledge must be severely limited, or the algorithm will overfit to the training games and perform poorly on the unknown games of the test set. Research in this area has been of special interest in the last years, with emerging contests like the General Video Game AI (GVG-AI) Competition. This paper introduces three different open loop techniques for dealing with this problem. First, a simple directed depth first search algorithm is employed as a baseline. Then, a tree search algorithm with a multi-armed bandit based tree policy is presented, followed by a Rolling Horizon Evolutionary Algorithm (RHEA) approach. In order to test these techniques, the games from the GVG-AI Competition framework are used as a benchmark, evaluation on a training set of 29 games, and submitting to the 10 unknown games at the competition website. Results show how the general game-independent heuristic proposed works well across all algorithms and games, and how the RHEA becomes the best evolutionary technique in the rankings of the test set.

MCTS/EA hybrid GVGAI players and game difficulty estimation

“…In a different approach [13], a RHEA builds a tree search while evaluating the different individuals. Every time a sequence of actions is executed from the current state, new nodes will be added to this tree, one per action performed.…”

Analysis of Vanilla Rolling Horizon Evolution Parameters in General Video Game Playing

“…These methods, rather than basing the search on game tree structures, use influences from biological sciences to evolve a population of individuals until a suitable one, corresponding to a solution to the problem, is obtained. The way they are applied to the domain of GVGP is by encoding sequences of in-game actions as individuals, using heuristics to analyse the value of each sequence [20].…”

General Video Game for 2 players: Framework and competition

“…Open Loop Monte Carlo Tree Search (OLMCTS) [12] is an MCTS variation, which does not store the game state in the nodes of the trees, instead using the forward model to reevaluate future states at each iteration. This method is particularly efficient in GVGAI due to the stochastic nature of the games, which means that storing game states in the nodes of the tree would result in possibly inaccurate results and wrong assumptions to be made based on previously different scenarios.…”