Comments (7)
So how does it have the structure? In pre-training, every step, a latent code is sampled randomly, so there is no guarantee that similar behaviors correspond to similar latent codes, is that right?
from ase.
Yes, the latents are sampled randomly during pre-training. Because of the objective used during pre-training, the model will learn to assign different behaviors to different latents automatically. This is similar to what happens in unsupervised reinforcement learning. We do not need to explicitly specify which skills a particular latent produces. Instead the GAN and unsupervised RL objective will automatically learn a skill embedding where different zs will be mapped to different behaviors that resemble the dataset. If you want more details, you can take a look at the paper for a more in-depth explanation.
from ase.
Yes, the latents are sampled randomly during pre-training. Because of the objective used during pre-training, the model will learn to assign different behaviors to different latents automatically. This is similar to what happens in unsupervised reinforcement learning. We do not need to explicitly specify which skills a particular latent produces. Instead the GAN and unsupervised RL objective will automatically learn a skill embedding where different zs will be mapped to different behaviors that resemble the dataset. If you want more details, you can take a look at the paper for a more in-depth explanation.
Thanks for your reply. Assuming the latent code is 1-dim(range from 0 to 1), for a specific skill, for example, 'jump', when performing 'jump' skill in high level stage, it should get continuously changing zs (0.1, 0.11, 0.12...)for sequential frames, right? because for the task policy network, the input is continuous, the output zs is also continuous. However, during pre-training, for 'jump' skill, I mentioned that for every 10 frames, the sequence amp_obs is mapped with a random zs, so the motion clip can be mapped to distinct zs(0.1, 0.5, 0.9), I think it may hard to generate 'jump' skill stably.
I guess an ideal map between zs and skill may be that a skill(motion clip) should map with a zs cluster, not discrete different zs.
hope for your opinion!
from ase.
sorry not sure if i understand your question.
from ase.
sorry i didn't ask clearly. What confused me is that after pre-training, is the latent space structured?Is it like the first picture or the second picture below? different color represents a specific skill.
from ase.
There is some structure in the latent space. Latents that are close in the latent space will typically correspond to similar behaviors.
from ase.
yes that's right. The latents are sampled randomly during training, and there's no guarantee that similar behaviors correspond to similar latent codes. But in practice, we do see that similar latent codes often lead to similar behaviors. This is likely partly due to the smoothness of the function approximator and the mutual information objective.
from ase.
Related Issues (20)
- Agent with many dofs: Discriminator accuracy suddenly drops to zero HOT 1
- About the DOF conversion of character motions. HOT 2
- Error when import FBX HOT 1
- How to tune hyper parameters for GPU's with more memory? HOT 4
- About the blend in the function "get_motion_state" HOT 2
- How to get 253D states from issacgym env
- ISAAC gym preview 4 requires rl-games>=1.6.0 and does not meet ASE program requirements HOT 3
- [fbx_sdk install manual] if you stucked in install fbx_sdk for import fbx or other files. check this HOT 1
- How do you come up with the "rotation" and "scale" value in the retarget config file? HOT 2
- Retargeted animation looks really weird (upperArm position)... Is that normal? HOT 3
- Characters are spawned in the sky? HOT 2
- Is there a Issac Sim version of this repo or something similar in the works? I just want to ask before attempting a port of this to Isaac sim.
- Installling requirements tips HOT 1
- tensors are created on cpu when cuda is specified as the device.
- sudden translation / rotation changes of motion loaded from fbx
- mocap data HOT 1
- The implementation seems to be different from the method in the paper HOT 1
- About the Implementation of FK
- More about motion weights
- the source of the CMU fbx files HOT 5
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from ase.