8000 Develop the dynamics model · Issue #89 · agi-brain/xuance · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Develop the dynamics model #89

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
yui-iii opened this issue Jan 13, 2025 · 1 comment
Open

Develop the dynamics model #89

yui-iii opened this issue Jan 13, 2025 · 1 comment
Milestone
Support Dreamer a...

Comments

@yui-iii
Copy link
yui-iii commented Jan 13, 2025

No description provided.

@wenzhangliu
Copy link
Collaborator

In Dreamer, the agent learns a latent world model that can be used to imagine or “dream” future trajectories without directly interacting with the environment. This model typically has three main parts:

  1. Representation Model / Encoder: Encodes the high-dimensional observation (e.g. image) into a lower-dimensional latent embedding.
  2. Dynamics Model / RSSM: Predicts stochastic transitions in the latent space, conditioned on actions.
  3. Decoder: Reconstructs the original observation and/or reward from the latent state (for training signals).

wenzhangliu added a commit that referenced this issue Apr 11, 2025
@wenzhangliu
DreamerV3 works in XuanCe (#89)
19298a2
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
Support Dreamer as a model-based algo...
Development

No branches or pull requests
2 participants
@wenzhangliu @yui-iii
0