Metraipolis › archive › dc_5a792eaa701e

Modeling and Understanding Neural Weights of Humanoid Policies with Diffusion — arXiv:2610.20606

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20606
License (as archived)
CC0-1.0
Archived
2026-10-15T14:23:57.857Z
SHA-256 of the archived text
3df9bcbd618ba64693234282ea1af3e0a42452a5593169e83b7d946ed66348fb
Stable URL of this page
https://metraipolis.com/documento/dc_5a792eaa701e

How to cite

Modeling and Understanding Neural Weights of Humanoid Policies with Diffusion — arXiv:2610.20606 — arXiv (en), https://arxiv.org/abs/2610.20606 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5a792eaa701e (Metraipolis archive dc_5a792eaa701e, sha256 3df9bcbd618ba64693234282ea1af3e0a42452a5593169e83b7d946ed66348fb)
BibTeX
@misc{dc5a792eaa701e,
  title = {Modeling and Understanding Neural Weights of Humanoid Policies with Diffusion — arXiv:2610.20606},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20606},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5a792eaa701e; sha256 3df9bcbd618ba64693234282ea1af3e0a42452a5593169e83b7d946ed66348fb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20606. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent advances in robot learning have driven rapid progress in humanoid control, enabling increasingly agile behaviors in the real world. However, humanoid control networks are typically obtained through conventional training procedures, while comparatively little attention has been given to directly modeling the distribution of their learned weights and generating functional networks in weight space. In this work, we investigate whether high-performing humanoid control networks can be generatively modeled directly through their weights. We adopt a latent diffusion framework that couples a linear autoencoder with a Diffusion Transformer. The autoencoder compresses actor network parameters into a compact latent space, and the Diffusion Transformer models the distribution of these latents. Trained on networks collected from reinforcement learning for the Unitree G1 robot, our approach generates deployable networks that perform various motions in the real world, including walking, dancing, and spin kick. Beyond generation, we further study the learned weight space to better understand the structure captured by these humanoid control networks. We identify a structural principle in the first-layer weights that consistently reflects the robot's kinematic prior. Notably, this kinematic prior emerges under both Reinforcement Learning (RL) and Imitation Learning (IL) training, suggesting that it is not specific to a particular optimization process but may instead reflect common knowledge associated with humanoid embodiment. Together, these results demonstrate the feasibility of generative modeling in humanoid network weight space, while providing initial insights into the information encoded by their learned weights. Additional results are available on our project website: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.20606 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T13:53:35.989Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).