An Introduction of mini-AlphaStar

TL;DR

This paper introduces mini-AlphaStar, a scaled-down version of the AlphaStar agent for StarCraft II, demonstrating its implementation and open-sourcing for future reinforcement learning research.

Contribution

The paper presents a mini-scale implementation of AlphaStar with adjusted hyperparameters, providing an accessible version for further research and experimentation.

Findings

Achieved a scaled-down version of AlphaStar for StarCraft II

Open-sourced the code for community use

Facilitates future reinforcement learning research in complex environments

Abstract

StarCraft II (SC2) is a real-time strategy game in which players produce and control multiple units to fight against opponent's units. Due to its difficulties, such as huge state space, various action space, a long time horizon, and imperfect information, SC2 has been a research hotspot in reinforcement learning. Recently, an agent called AlphaStar (AS) has been proposed, which shows good performance, obtaining a high win rate of 99.8% against human players. We implemented a mini-scaled version of it called mini-AlphaStar (mAS) based on AS's paper and pseudocode. The difference between AS and mAS is that we substituted the hyper-parameters of AS with smaller ones for mini-scale training. Codes of mAS are all open-sourced (https://github.com/liuruoze/mini-AlphaStar) for future research.