swMATH ID: 38437
Software Authors: Michał Kempka, Marek Wydmuch, Grzegorz Runc, Jakub Toczek, Wojciech Jaśkowski
Description: ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning. The recent advances in deep neural networks have led to effective vision-based reinforcement learning methods that have been employed to obtain human-level controllers in Atari 2600 games from pixel data. Atari 2600 games, however, do not resemble real-world tasks since they involve non-realistic 2D environments and the third-person perspective. Here, we propose a novel test-bed platform for reinforcement learning research from raw visual information which employs the first-person perspective in a semi-realistic 3D world. The software, called ViZDoom, is based on the classical first-person shooter video game, Doom. It allows developing bots that play the game using the screen buffer. ViZDoom is lightweight, fast, and highly customizable via a convenient mechanism of user scenarios. In the experimental part, we test the environment by trying to learn bots for two scenarios: a basic move-and-shoot task and a more complex maze-navigation problem. Using convolutional deep neural networks with Q-learning and experience replay, for both scenarios, we were able to train competent bots, which exhibit human-like behaviors. The results confirm the utility of ViZDoom as an AI research platform and imply that visual reinforcement learning in 3D realistic first-person perspective environments is feasible.
Homepage: https://arxiv.org/abs/1605.02097
Keywords: Machine Learning; arXiv_cs.LG; Artificial Intelligence; arXiv_cs.AI; Computer Vision; Pattern Recognition; arXiv_cs.CV
Related Software: MuJoCo; AI2-THOR; ImageNet; Matterport3D; VECA; VIME; IMPALA; PILCO; pix2pix; CycleGAN; Adam; OpenAI Gym; darch; Theano
Referenced in: 2 Publications

Referencing Publications by Year