Clearly needs some energy constraint factored in to look less like monty python :) I also wonder if the gravity is realistic.<p>A similar video from 2013: <a href="https://www.youtube.com/watch?v=pgaEE27nsQw" rel="nofollow">https://www.youtube.com/watch?v=pgaEE27nsQw</a>
Looks much smoother, because of muscle and nerve simulations. At 0:55 you can see the generations learning how to walk.
If anyone is interested in learning more how reinforcement learning works, I would highly recommend openai gym [0]. It provides a controlled environment where you can develop learning algorithms to accomplish an objective goal. It also has a walkers similar to the one in the video [1]. For learning more about reinforcement learning, I would recommend Richard S. Sutton book Reinforcement Learning: An Introduction. He has a new draft available free online [2]. Or the course Practical RL [3].<p>[0] <a href="https://gym.openai.com/envs" rel="nofollow">https://gym.openai.com/envs</a><p>[1] <a href="https://gym.openai.com/envs#mujoco" rel="nofollow">https://gym.openai.com/envs#mujoco</a><p>[2] <a href="http://ufal.mff.cuni.cz/~straka/courses/npfl114/2016/sutton-bookdraft2016sep.pdf" rel="nofollow">http://ufal.mff.cuni.cz/~straka/courses/npfl114/2016/sutton-...</a><p>[3] <a href="https://github.com/yandexdataschool/Practical_RL" rel="nofollow">https://github.com/yandexdataschool/Practical_RL</a>
We are pretty good at walking and moving fast forward..<p>I wonder how fast a AI could get at climbing- walking alterations, when given a challenging terrain..
one more sensationalist headline for deepmind<p><a href="https://www.google.es/search?q=mujoco+learning+to+walk#q=mujoco+learning+to+walk&tbm=vid" rel="nofollow">https://www.google.es/search?q=mujoco+learning+to+walk#q=muj...</a>