Deep Reinforcement Learning for Playing 2.5D Fighting Games
This work addresses a domain-specific problem for game AI researchers, but it appears incremental as it builds on existing A3C methods with modifications for a particular game type.
The paper tackled the challenge of applying deep reinforcement learning to 2.5D fighting games, which involve visual ambiguities and complex sequential actions, by proposing an A3C+ network with a Recurrent Info component and successfully demonstrated its use in the game Little Fighter 2.
Deep reinforcement learning has shown its success in game playing. However, 2.5D fighting games would be a challenging task to handle due to ambiguity in visual appearances like height or depth of the characters. Moreover, actions in such games typically involve particular sequential action orders, which also makes the network design very difficult. Based on the network of Asynchronous Advantage Actor-Critic (A3C), we create an OpenAI-gym-like gaming environment with the game of Little Fighter 2 (LF2), and present a novel A3C+ network for learning RL agents. The introduced model includes a Recurrent Info network, which utilizes game-related info features with recurrent layers to observe combo skills for fighting. In the experiments, we consider LF2 in different settings, which successfully demonstrates the use of our proposed model for learning 2.5D fighting games.