论文标题
实时行动策略游戏的深度RL代理
Deep RL Agent for a Real-Time Action Strategy Game
论文作者
论文摘要
我们介绍了一个基于英雄 - 魔术对决（1 V 1动作策略游戏）的强化学习环境。该域是无处不在的，原因有几个：这是一个实时游戏，状态空间很大，在比赛的每个步骤中给出的信息不完善，并且动作的分布是动态的。我们的主要贡献是一位深厚的强化学习代理商以竞争性的水平在游戏水平上玩游戏，我们使用PPO和自我玩法与多个竞争代理商进行了训练，仅采用$ \ pm 1 $的简单奖励，具体取决于单场比赛的结果。我们最好的自我游戏代理商对现有AI的$ 65 \％$获胜率和超过$ 50 \％$ $ $ $ $ $ $。
We introduce a reinforcement learning environment based on Heroic - Magic Duel, a 1 v 1 action strategy game. This domain is non-trivial for several reasons: it is a real-time game, the state space is large, the information given to the player before and at each step of a match is imperfect, and distribution of actions is dynamic. Our main contribution is a deep reinforcement learning agent playing the game at a competitive level that we trained using PPO and self-play with multiple competing agents, employing only a simple reward of $\pm 1$ depending on the outcome of a single match. Our best self-play agent, obtains around $65\%$ win rate against the existing AI and over $50\%$ win rate against a top human player.
