Corpus ID: 28808621

StarCraft II: A New Challenge for Reinforcement Learning

  title={StarCraft II: A New Challenge for Reinforcement Learning},
  author={Oriol Vinyals and Timo Ewalds and Sergey Bartunov and P. Georgiev and A. S. Vezhnevets and Michelle Yeo and Alireza Makhzani and Heinrich K{\"u}ttler and J. Agapiou and Julian Schrittwieser and John Quan and Stephen Gaffney and S. Petersen and K. Simonyan and T. Schaul and H. V. Hasselt and D. Silver and T. Lillicrap and Kevin Calderone and Paul Keet and Anthony Brunasso and D. Lawrence and Anders Ekermo and J. Repp and Rodney Tsing},
  • Oriol Vinyals, Timo Ewalds, +22 authors Rodney Tsing
  • Published 2017
  • Computer Science
  • ArXiv
  • This paper introduces SC2LE (StarCraft II Learning Environment), a reinforcement learning environment based on the StarCraft II game. [...] Key Method We describe the observation, action, and reward specification for the StarCraft II domain and provide an open source Python-based interface for communicating with the game engine. In addition to the main game maps, we provide a suite of mini-games focusing on different elements of StarCraft II gameplay. For the main game maps, we also provide an accompanying…Expand Abstract
    385 Citations
    On Reinforcement Learning for Full-length Game of StarCraft
    • 30
    • PDF
    Efficient Reinforcement Learning with a Thought-Game for StarCraft
    • 1
    • Highly Influenced
    SCC: an efficient deep reinforcement learning agent mastering the game of StarCraft II
    • PDF
    A Hierarchical Model for StarCraft II Mini-Game
    • Highly Influenced
    Grandmaster level in StarCraft II using multi-agent reinforcement learning
    • 512
    • PDF
    MinAtar: An Atari-Inspired Testbed for Thorough and Reproducible Reinforcement Learning Experiments
    • 9
    Deep Reinforcement Learning in Starcraft II
    • Highly Influenced
    • PDF


    The Arcade Learning Environment: An Evaluation Platform for General Agents (Extended Abstract)
    • 1,485
    • Highly Influential
    • PDF
    ViZDoom: A Doom-based AI research platform for visual reinforcement learning
    • 440
    • PDF
    Multiagent Bidirectionally-Coordinated Nets for Learning to Play StarCraft Combat Games
    • 137
    • PDF
    Playing SNES in the Retro Learning Environment
    • 13
    • PDF
    Learning macromanagement in starcraft from replays using deep learning
    • Niels Justesen, S. Risi
    • Computer Science
    • 2017 IEEE Conference on Computational Intelligence and Games (CIG)
    • 2017
    • 48
    • PDF
    Learning from Demonstrations for Real World Reinforcement Learning
    • 109
    • PDF
    Deep Q-learning From Demonstrations
    • 366
    • PDF
    Mastering the game of Go with deep neural networks and tree search
    • 8,244
    • PDF