A Dataset for StarCraft AI & an Example of Armies Clustering
By Gabriel Synnaeve, Pierre Bessière. 2012
44 citations recorded by the source corpus.
This paper advocates the exploration of the full state of recorded real-time strategy (RTS) games, by human or robotic players, to discover how to reason about tactics and strategy. We present a dataset of StarCraft games encompassing the most of the games' state (not only player's orders). We explain one of the possible usages of this dataset by clustering armies on their compositions. This reduction of armies compositions to mixtures of Gaussian allow for strategic reasoning at the level of the components. We evaluated this clustering method by predicting the outcomes of battles based on armies compositions' mixtures components
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Yoshua Bengio
- Nantas Nardelli
- Nicolas Usunier
- Niels Justesen
- Sebastian Risi
- Timothée Lacroix
- Zeming Lin
- Quoc V. Le
Citations in the corpus
- TorchCraft: a Library for Machine Learning Research on Real-Time Strategy Games (2016, methodology): “Other models are trained on offline datasets of highly skilled players [25, 18, 19, 21].”
- Global State Evaluation in StarCraft (2014, methodology): “Our technique for identifying battles, as shown by Algorithm 1, is very similar to one presented in (Synnaeve and Bessiere 2012).”
- Learning macromanagement in starcraft from replays using deep learning (2017, background): “collected a repository of 7,649 replays by scraping the three StarCraft community websites GosuGamers, ICCup and TeamLiquid, which are mainly for highly skilled players including professionals [27].”
- A Survey of Real-Time Strategy Game AI Research and Competition in StarCraft (2013, unknown): “One the other hand, large datasets of replays have been created [4], [5], from where strategies, trends or plans have been tried to learn.”