Quan Vuong (@QuanVng) / X

Quan Vuong

472 posts

Quan Vuong

@QuanVng

robotics research and co-founder at @Physical_int, ex-@GoogleDeepMind Perpetually trying to find a quiet place to read.

linkedin.com/in/quan-vuong-…

Joined January 2015

Pinned
Quan Vuong
@QuanVng
Apr 24, 2025
And we're hiring! Apply on our website or DM me with questions
Physical Intelligence
@physical_int
Apr 22, 2025
We got a robot to clean up homes that were never seen in its training data! Our new model, π-0.5, aims to tackle open-world generalization. We took our robot into homes that were not in the training data and asked it to clean kitchens and bedrooms. More below⤵️
00:00
22K
Quan Vuong
@QuanVng
Oct 3, 2023
RT-X: generalist AI models lead to 50% improvement over RT-1 and 3x improvement over RT-2, our previous best models. 🔥🥳🧵 Project website: robotics-transformer-x.github.io
GIF
337K
Quan Vuong
@QuanVng
Dec 25, 2024
I'm hiring exceptional researchers, engineers (both research and full-stack) at @physical_int. Please apply on our website or DM me with questions. Referrals very appreciated!
35K
Quan Vuong
@QuanVng
Feb 4, 2025
And we're also hiring! apply directly on our website or dm me with questions!
Physical Intelligence
@physical_int
Feb 4, 2025
Many of you asked for code & weights for π₀, we are happy to announce that we are releasing π₀ and pre-trained checkpoints in our new openpi repository! We tested the model on a few public robots, and we include code for you to fine-tune it yourself.
00:00
8.2K
Quan Vuong
@QuanVng
Jul 23, 2024
O-1A approved! Thanks to @minney_cat and @lighthousehq_ for their guidance and support throughout the whole process. Would recommend working with them for your visa needs. Blazingly fast, no fuss and very responsive!
12K
Quan Vuong
@QuanVng
Mar 12, 2024
Karol Hausman
@hausman_k
Mar 12, 2024
🚨 Big news 🚨 Together with a set of amazing folks we decided to start a company that tackles one of the hardest and most impactful problems - Physical Intelligence In fact, we even named our company after that: physicalintelligence.company or Pi (π) for short 🧵
Physical Intelligence (π)
From pi.website
14K
Quan Vuong
@QuanVng
May 17, 2024
Large, ambitious projects in robotics are fun and really worthwhile We should do more of them!
3.8K
Quan Vuong
@QuanVng
Nov 9, 2023
Pictures taken at RT-2 poster at @DannyDriess requests ; ) @YevgenChebotar We miss you @TianheYu CC @hausman_k
8.2K
Quan Vuong
@QuanVng
Dec 11, 2019
The code for optimistic actor critic is now open-sourced.
Microsoft Research
@MSFTResearch
Nov 26, 2019
Optimistic Actor Critic, with the principle of optimism in the face of uncertainty, obtains an exploration policy by using the upper bound instead of the lower bound. Learn how OAC increases sample efficiency compared to other methods: aka.ms/AA6pk3q #NeurIPS2019
github.com
GitHub - microsoft/oac-explore: Code accompanying the paper "Better Exploration with Optimistic...
Code accompanying the paper "Better Exploration with Optimistic Actor Critic" (NeurIPS 2019) - microsoft/oac-explore
Quan Vuong
@QuanVng
Jan 4, 2024
Co-training with static data improves perf on mobile tasks. Signs of generalist policies to come : )
Tony Zhao
@tonyzzhao
Jan 3, 2024
Introducing 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 -- Hardware! A low-cost, open-source, mobile manipulator. One of the most high-effort projects in my past 5yrs! Not possible without co-lead @zipengfu and @chelseabfinn. At the end, what's better than cooking yourself a meal with the 🤖🧑‍🍳
00:00
7K
Quan Vuong
@QuanVng
Mar 24, 2019
Tried mathpix to convert picture into latex and I was blown away by how well it worked 😍
Mathpix
@MathpixApp
Feb 26, 2019
No spaces? No problem! New features just released on Mac, @Windows, and @ubuntu Snip! mathpix.com
GIF
Quan Vuong
@QuanVng
Mar 28, 2019
Mind-blown from reading this paper: meta-gradient reinforcement learning (arxiv.org/abs/1805.09801). A learning algorithm which learns to edit itself online during the training process at no extra data cost. Online cross-validation is so cool 🤯🤯🤯
arxiv.org
Meta-Gradient Reinforcement Learning
The goal of reinforcement learning algorithms is to estimate and/or optimise the value function. However, unlike supervised learning, no teacher or oracle is available to provide the true value...
Quan Vuong
@QuanVng
Dec 15, 2023
@KarlPertsch and I are giving a talk about RT-X future plans at Room 354 at 3PM, come discuss with us! #NeurIPS23
20K
Quan Vuong
@QuanVng
Oct 29, 2019
1. Optimistic Actor Critic (neurips 2019 spotlight) Existing tricks to stabilize training leads to pessimistic exploration. We introduce optimistic exploration and obtain sample efficiency gains! Paper: arxiv.org/abs/1910.12807
Quan Vuong
@QuanVng
Oct 29, 2019
Excited to share 3 recent works!