Forward Modeling for Partial Observation Strategy Games - A StarCraft Defogger

By Gabriel Synnaeve, Zeming Lin, Jonas Gehring, Daniel Gant, Vegard Mella, Vasil Khalidov, Nicolas Carion, Nicolas Usunier. 2018

27 citations recorded by the source corpus.

We formulate the problem of defogging as state estimation and future state prediction from previous, partial observations in the context of real-time strategy games. We propose to employ encoder-decoder neural networks for this task, and introduce proxy tasks and baselines for evaluation to assess their ability of capturing basic game rules and high-level dynamics. By combining convolutional neural networks and recurrent networks, we exploit spatial and sequential correlations and train well-performing models on a large dataset of human games of StarCraft: Brood War. Finally, we demonstrate the relevance of our models to downstream tasks by applying them for enemy unit prediction in a state-of-the-art, rule-based StarCraft bot. We observe improvements in win rates against several strong community bots.

Subfields: Artificial Intelligence, Sociology and Political Science, Reinforcement Learning, Signal Processing

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

1. Sergey Levine
2. Yoshua Bengio
3. Pieter Abbeel
4. Mohit Bansal
5. Ruslan Salakhutdinov
6. Quoc V. Le
7. Kyunghyun Cho
8. Jianfeng Gao

In Sociology and Political Science

1. Aja Huang
2. Zeming Lin
3. Adam Lerer
4. Junhyuk Oh
5. Wojciech M. Czarnecki
6. David Silver
7. Dani Yogatama
8. Soumith Chintala

In Reinforcement Learning

1. Sergey Levine
2. Pieter Abbeel
3. Yuxi Li
4. Timothy P. Lillicrap
5. David Silver
6. Shimon Whiteson
7. Marc G. Bellemare
8. Nicolas Manfred Otto Heess

In Signal Processing

1. Oriol Vinyals
2. Yoshua Bengio
3. Wojciech M. Czarnecki
4. Caglar Gulcehre
5. Koray Kavukcuoglu
6. Junyoung Chung
7. Timothy P. Lillicrap
8. James Bradbury

Citations in the corpus

• PyTorch: An Imperative Style, High-Performance Deep Learning Library (2019, methodology): “This design also allowed us to create first-class C++ bindings and modeling libraries that can be used in places where Python is inconvenient, such as the game engine for Starcraft [36] or on mobile platforms.”
• Grandmaster level in StarCraft II using multi-agent reinforcement learning (2019, methodology): “Most recently, deep learning has been used to predict future game states 46 .”