Reinforcement learning with musculoskeletal models

Re:Reinforcement learning with musculoskeletal models by organicmultiloc on Thursday May 17, 2018 @12:13PM (#17092621)

I may be late to the game but this is the first time I've looked at something and thought "yeah well The Terminator is about to happen"

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by tedivm on Thursday May 17, 2018 @12:41PM (#17092928)

The skeleton is not exactly good marketing.

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by stcredzero on Thursday May 17, 2018 @01:22PM (#17093315)

I was going to suggest that they hook the AI up to QWOP.

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by tlrobinson on Thursday May 17, 2018 @03:51PM (#17094909)

I was curious if anyone had tried to train an AI to play QWOP. Of course they have: https://www.youtube.com/watch?v=e27TUmMkOA0 (among others)

I wonder if you could get a better result by including other factors in the reward function, like trying to maintain a slight forward lean.

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by mxwsn on Thursday May 17, 2018 @02:19PM (#17093931)

But this task is strictly harder and more general than QWOP .

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by juancb on Thursday May 17, 2018 @02:59PM (#17094397)

But not strictly as humourous.

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by stcredzero on Thursday May 17, 2018 @03:49PM (#17094882)

For PR purposes.

[ Parent | on HN ]

Re:Reinforcement learning with musculoskeletal models by mooneater on Thursday May 17, 2018 @01:08PM (#17093168)

there is a huge chasm between this working in sim, and irl.

[ Parent | on HN ]