Reinforcement learning with musculoskeletal models

roel_v 8 years ago |

Like literally 3 hours ago I was searching for software that can simulate skeletons and joints to investigate the exact effects of jiu jitsu joint locks, optimal fulcrum points etc. Could this be used or adapted for that? Anyone here ever used it?

altvali 8 years ago | |

I've participated in the NIPS "Learning to run" competition last year. It only computes motion in one direction (forward-backward, the model can't go left or right). What turned me off was the fact that object collision was handled poorly. You can't see a model that actually makes effort to avoid obstacles, instead all top solutions go through objects.

kidzik 8 years ago | | |

That's only one-side of the story. It is actually a very accurate contact model. The problem is that it's also computationally expensive and therefore we reduce the stiffness of objects to make it faster. It's basically a trade-off between the accuracy and speed. In gaming engines you have speed and it looks good, but then it's impossible to generalize for real-life applications (because the contact is inaccurate).

electricslpnsld 8 years ago | | |

> The problem is that it's also computationally expensive and therefore we reduce the stiffness of objects to make it faster.

Are y'all using penalty methods for the collisions? Which model does it use?

roel_v 8 years ago | | |

So all 'joints' only move in 2d, am I getting that right?

kidzik 8 years ago | | |

In the 2017 challenge yes. This year we are using a 3d model.

nurettin 8 years ago | |

I know a guy who is interested in that, take a look at this: https://github.com/Eelis/GrappleMap/blob/master/README.md

roel_v 8 years ago | | |

That's really cool, thanks for mentioning.

mr_overalls 8 years ago | |

Haven't used the software, but how about pitting intelligent agents against each other (incorporating reasonable models for vision, motor control, etc.) and evolving the optimal martial art? :-)

jcims 8 years ago | |

Might find this interesting for extracting info from recorded bouts: https://github.com/CMU-Perceptual-Computing-Lab/openpose

chatmasta 8 years ago | |

I’ve also been searching for something like this, wondering if it could be applied to horse racing simulation.

kidzik 8 years ago | | |

Absolutely, however, you will need to construct a muscle model for horses. The human model is a result of decades of research, MRIs etc.

adamnemecek 8 years ago |

Anyone into this check out dual quaternions

http://www.chinedufn.com/dual-quaternion-shader-explained/

They are magic.

budadre75 8 years ago |

I tried the 2017 competition, and DDPG never converges for me. Since the 30+ dimensional state space is so large, I wonder are there some ML techniques that deal with dimensionality reduction or just large dimensions in general?

jbattle 8 years ago |

A lot of these would be good submissions to the Ministry of Silly Walks

https://www.youtube.com/watch?v=iV2ViNJFZC8

pixelHD 8 years ago |

Wow, would be nice if this was integrated into the OpenAI environments

mooneater 8 years ago | |

if you look at the github readme, it is openai gym compatible.

organicmultiloc 8 years ago |

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"

tedivm 8 years ago | |

The skeleton is not exactly good marketing.

stcredzero 8 years ago | | |

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

tlrobinson 8 years ago | | |

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.

mxwsn 8 years ago | | |

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

juancb 8 years ago | | |

But not strictly as humourous.

stcredzero 8 years ago | | |

For PR purposes.

mooneater 8 years ago | |

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

sudouser 8 years ago |

can this be used to help children with developmental delays and disabilities?