Learning to Win: Case-Based Plan Selection in a Real-Time Strategy Game

By David William Aha, Matthew Molineaux, Marc J. V. Ponsen. 2005

222 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

TorchCraft: a Library for Machine Learning Research on Real-Time Strategy Games (2016, background): “Real-time strategy (RTS) games have historically been a domain of interest of the planning and decision making research communities [5, 2, 6, 16, 17].”
A Dataset for StarCraft AI & an Example of Armies Clustering (2012, unknown): “Case-based reasoning (CBR) approaches often use extensions of build trees as state lattices (and sets of tactics for each state) as for (Aha, Molineaux, and Ponsen 2005; Ponsen and Spronck 2004) in Wargus.”
Heuristic Search Techniques for Real-Time Strategy Games (2016, methodology): “[1] also working in the Wargus domain used case-based reasoning (CBR) for dynamic plan retrieval in their Case-based Tactician (CaT) system which was successful in defeating scripted and evolved opponents.”
A data mining approach to strategy prediction (2009, methodology): “Reinforcement learning [9] and Monte Carlo [10] methods have shown promising results for tactical aspects in RTS games, while genetic algorithms [11] and case-based reasoning [12] have been applied to strategy selection.”
A Review of Real-Time Strategy Game AI (2014, methodology): “A drawback of this method is that the effectiveness values learned for the dynamic scripts assumes that the opponent is static and would not adapt well to a dynamic opponent (Aha, Molineaux, and Ponsen 2005).”
Programmation et apprentissage bayésien pour les jeux vidéo multi-joueurs, application à l'intelligence artificielle de jeux de stratégies temps-réel. (Bayesian Programming and Learning for Multi-Player Video Games, Application to RTS AI) (2012, methodology): “103 [Aha et al., 2005] (Chess: ⇡ 35, Go: < 360): Weber [2012] finds a branching factor greater than 1.”
A Bayesian model for opening prediction in RTS games with application to StarCraft (2011, methodology): “We used this replay feature to extract players actions and learn the probabilities of tech trees to happen at a given time and, in this paper, also given a labeled opening.”
Case-Based Planning and Execution for Real-Time Strategy Games (2007, methodology): “Speciﬁcally, we are interested in real-time strategy (RTS) games, that have been shown to have huge decision spaces that cannot be dealt with search based AI techniques [2, 3].”
A Survey of Real-Time Strategy Game AI Research and Competition in StarCraft (2013, methodology): “[18] used CBR to perform dynamic plan retrieval in the Wargus domain.”
A Bayesian model for RTS units control applied to StarCraft (2011, methodology): “Some interesting uses of RL [16] to RTS research are concurrent hierarchical (units Q-functions are combined higher up) RL [5] to efﬁciently control units in a multi-effector system fashion, and large-scale strategy games [6].”
Behavior Trees in Robotics and AI: An Introduction (2017, background): “Building agents that perform well in this domain is challenging due to the large decision space [1].”