Reinforcement learning combined with deep neural networks has performed remarkably well in many genres of games recently. It has surpassed human-level performance in fixed game environments and turn-based two player board games. However, to the best of our knowledge, current research has yet to produce a result that has surpassed human-level performance in modern complex fighting games. This is due to the inherent difficulties with real-time fighting games, including: vast action spaces, action dependencies, and imperfect information. We overcame these challenges and made 1v1 battle AI agents for the commercial game "Blade & Soul". The trained agents competed against five professional gamers and achieved a win rate of 62%. This paper presents a practical reinforcement learning method that includes a novel self-play curriculum and data skipping techniques. Through the curriculum, three different styles of agents were created by reward shaping and were trained against each other. Additionally, this paper suggests data skipping techniques that could increase data efficiency and facilitate explorations in vast spaces. Since our method can be generally applied to all two-player competitive games with vast action spaces, we anticipate its application to game development including level design and automated balancing.

No abstract

Unlike the situation with regard to board games, artificial intelligence (AI) for real-time strategy (RTS) games usually suffers from an infinite number of possible future states.Furthermore, it must handle the complexity quickly. This constraint makes it difficult to build AI for RTS games with current state-of-the-art intelligent techniques. This paper proposes the use of imitation learning based on a human player's replays, which allows the AI to mimic the behaviors. During game play, the AI exploits the replay repository to search for the best similar moment from an influence map representation. This work focuses on combat in RTS games, considering the spatial configuration and unit types.Experimental results show that the proposed AI can defeat well-known competition entries a large percentage of the time.

No abstract