Google DeepMind AI Does Parkour - Producing flexible behaviours in simulated environments

Google's AI company DeepMind created an AI that has learned how to walk... or how to parkour without any prior instructions. Emergence of Locomotion Behaviours in Rich Environments. Read the blog post here: https://deepmind.com/blog/producing-flexible-behaviours-simulated-environments/ Research performed by: Nicolas Heess, Dhruva TB, Srinivasan Sriram, Jay Lemmon, Josh Merel, Greg Wayne, Yuval Tassa, Tom Erez, Ziyu Wang, Ali Eslami, Martin Riedmiller, David Silver Via: https://arxiv.org/abs/1707.02286

• #science_technology
• #google
• #machine_learning
• #ai
• #deepmind
• #parkour
• #ai
• #deepmind_parkour
• #emergence_of_locomotion
• #deep_mind_parkour
• #google_deepmind_animation
• #tech_insider
• #google_ai
• #google_ai_animation
• #google_deepmind_teaches_itself_to_walk
• #walking_google_deepmind
• #googles_deepmind_ai_just_taught_itself_to_walk
• #learn_how_to_walk
• #ai_learns_to_walk
• #simulated_walking
• #simulated_ai
• #google_simulation
• #google_ai_simulation
• #google_deepmind_research