Skip to content
View tactino's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Organizations

@PlugRL

Block or report tactino

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
tactino/README.md

Zuo Gou

B.Eng. in Automation at Tsinghua University. I work on embodied AI and robot learning — how an agent acquires, shares and acts on the information it needs to move through the physical world.

What I'm working on

Shared perception between humanoids · Research intern, Prof. Wei-Chiu Ma's group, Cornell Two humanoids carrying an object together occlude exactly what each other needs to see. I study how they should share perception over a lossy link — what to send, how much, how often — and how the receiver should behave when the link drops. Humanoid2Humanoid, under review.

Noise schedules for joint continuous–discrete diffusion · Research intern, Prof. Ge Liu's group, UIUC When a diffusion model generates continuous and discrete quantities at once — atom positions alongside atom types — how should noise be paced across the two channels? Currently hand-tuned per dataset; I work on when that pacing matters and what the best schedule looks like.

PlugRL · Research assistant, Prof. Huazhe Xu's group, IIIS, Tsinghua A written boundary between RL training and its environments, so that either side can be replaced without touching the other. A full-size π₀.₅ has run end to end through it on LIBERO.

Interests

Multi-robot perception and communication · manipulation and locomotion policies · diffusion and flow models for control · the measurement problem in all of the above — making sure a result says what you think it says.

Elsewhere

Homepage · CV · gouz23 [at] mails.tsinghua.edu.cn

Pinned Loading

  1. PlugRL/plugrl-protocol PlugRL/plugrl-protocol Public

    The written boundary between RL training and its environments: WebSocket + msgpack, with a feedback return channel. SPEC.md is normative and its checkable clauses are tests; two reference env clien…

    Python

  2. PlugRL/plugrl-server PlugRL/plugrl-server Public

    Training server for PlugRL - policy and algorithm behind a WebSocket boundary with a feedback return channel. A full-size pi0.5 ran end to end through it on LIBERO: inference, feedback and FPO trai…

    Python

  3. eai-project-lerobot eai-project-lerobot Public

    Diffusion Policy for the SO-101 dual-arm manipulator in SAPIEN, covering Lift, Stack and Sort tasks — data collection, Hydra-configured training and evaluation.

    Python 2

  4. so101-grasp-cube so101-grasp-cube Public

    Training and deploying ACT policies for SO-101 cube grasping: motion-planning data collection, ManiSkill-to-LeRobot conversion, then simulation and real-robot evaluation.

    Python 1

  5. eai-assignments eai-assignments Public

    Embodied AI coursework: 3D Gaussian Splatting, a robot simulation and RL maze-navigation pipeline, and model-based RL.

    Python 1

  6. pr-ml-assignments pr-ml-assignments Public

    Pattern Recognition and Machine Learning coursework — SVM, decision tree and perceptron implemented from scratch, each with experiments and a LaTeX report.

    TeX 1