TorchCraft: a Library for Machine Learning Research on Real-Time Strategy Games
This work addresses the problem of creating accessible tools for AI researchers to study complex, real-time decision-making in games, though it is incremental as it builds on existing game interfaces and frameworks.
The authors tackled the challenge of enabling deep learning research on Real-Time Strategy (RTS) games by developing TorchCraft, a library that facilitates control of games like StarCraft: Brood War from the Torch machine learning framework, and argued for using RTS games as an AI benchmark.
We present TorchCraft, a library that enables deep learning research on Real-Time Strategy (RTS) games such as StarCraft: Brood War, by making it easier to control these games from a machine learning framework, here Torch. This white paper argues for using RTS games as a benchmark for AI research, and describes the design and components of TorchCraft.