Activity
Ratings Progression
Challenge Categories
Challenges Entered
Completed
Stanford Neuromuscular Biomechanics Laboratory
Reinforcement Learning on Musculoskeletal Models
Latest submissions
See All| graded | 20188 | ||
| graded | 19710 | ||
| graded | 19677 |
| Participant | Rating |
|---|
| Participant | Rating |
|---|
NeurIPS 2019: Learn to Move - Walk Around
Questions about evaluation set up
Over 4 years ago@andrey_zubkov: Did you get to figure out the issue? 52 steps means that the human model fell down in 0.52 s and total reward of 5 seems about right in such a case. The difference between your local environment and the server environment can be on the initial state where the muscle states can be slightly different when a simulation is initiated by env.reset(...). So your controller should be robust enough to overcome the difference in init muscle state.
In addition, based on reports from other participants, it seems like there could have been more differences in the server evaluation. For Round 2, we only accept docker submission. So please try a docker submission and let us know if you still get drastically different results.
Google Cloud Credit - Round 2
Over 4 years ago@student Yes, you should’ve received an email with google cloud credit codes and further instructions a couple of days ago. Please let me know if you haven’t.
.reset() creates a new 'Figure' window each time
Over 4 years agoThis issue is solved in the new version: https://github.com/stanfordnmbl/osim-rl/commit/dfb9ad48721bd93ed31a0eba876e29ed8c7d88ed
Different reward on local and remote environments
Over 4 years agoSorry for the inconvenience. We will make sure the environments are consistent in Round 2.
Reward function for Round 2
Over 4 years agoThis thread is to discuss the reward for Round 2, the final round. Hopefully, the discussion will help us to set a good reward by Oct 14, when Round 2 begins. Before sharing your concerns or asking for clarifications, please check out the following links:
Raw observation usage – Osim
Over 4 years agoHi @jbr. Try obs_dict, reward, done, info = env.step([0] * 22, project=False)
Raw observation usage – Osim
Over 4 years ago@scitator Anything is allowed for your training, but the model=3D, difficulty=2, project=True, and obs_as_dict=True (i.e. the projection) will be used in the evaluation of Round 1.
Observation Dictionary to List Function
Over 4 years agoFYI, self.LENGTH0 is there to normalize the velocities to model size (i.e. leg length) but is not relevant for the competition as it is set to self.LENGTH0 = 1.
Clarification about Google cloud credits
Over 4 years ago@AlexanderKoch Please email me (smsong@stanford.edu) with your code, and I will send you a new one.
Clarification about Google cloud credits
Over 4 years agoSorry for the delay. We are trying to send out the credits every weekend to those who newly appear in the ranking, but there seems to be some delay. We will send out those shortly. Please state in this thread when you receive them. Thanks.
Questions about evaluation set up
Over 4 years agoHi, @luisenp. The evaluation in Round 1 will be done with the same environment with the last version of the development environment. When you submit your solution, our server runs multiple simulations with different target velocity fields, and the evaluation score is the mean of the cumulative rewards you receive in those simulations.
How can i get the 97D body state in the evaluation environment?
Over 4 years ago@gupengju The problem should’ve been solved (let us know if it not). Also, note that now you have another way to submit your solution: https://www.aicrowd.com/organizers/stanford-neuromuscular-biomechanics-laboratory/challenges/neurips-2019-learn-to-move-walk-around#get-started
Velocity target explanation
Over 4 years ago@huixxi You can find a new training example here: http://osim-rl.stanford.edu/docs/nips2019/training/
Google support for the top 200?
Over 4 years agoWe will send out the $200 credit codes to those who qualified shortly 
How can i get the 97D body state in the evaluation environment?
Over 4 years agoThe evaluation environment will be with the dictionary with 4 keys.
Velocity target explanation
Almost 5 years ago@huixxi: Sorry for the confusion. You are correct and thanks for pointing it out. I corrected the document (http://osim-rl.stanford.edu/docs/nips2019/environment/#observation-input-to-your-controller)
How can i get the 97D body state in the evaluation environment?
Almost 5 years agoSorry about the confusion. We recently noticed this issue and am working on it so that the evaluation environment will give the same observation dictionary as the current local environment. We will let you know once this is solved.
Modification of Environment Physics
Almost 5 years agoYou can change the physics parameters in the .osim file. For example, you can set zero gravity in the ./osim/models/gait14dof22musc_20170320.osim by changing line 9 to 0 0 0. Hope this helps.
Call for collaboration
Almost 5 years agoWe encourage participants with different background to collaborate and try interdisciplinary and novel approaches. Use this thread to call and search partners. For example, you can post something like this:
[Looking for deep RL experts]
Out current team: two gait biomechanics experts (“name 1” and “name 2”; homepage)
Goal: to win the “Most novel biomechanics solution” track
Contact: Email your@email.com to discuss more details
Questions about evaluation set up
Over 4 years ago@andrey_zubkov Try the docker submission for Round 2 and let us know if you have the issue.