Jonathan Tompson @JonathanTompson
Research Scientist at Google Joined November 2014-
Tweets17
-
Followers282
-
Following31
-
Likes59
Introducing 𝐀𝐋𝐎𝐇𝐀 𝐔𝐧𝐥𝐞𝐚𝐬𝐡𝐞𝐝 🌋 - Pushing the boundaries of dexterity with low-cost robots and AI. @GoogleDeepMind Finally got to share some videos after a few months. Robots are fully autonomous filmed in one continuous shot. Enjoy!
For the past year we've been working on ALOHA Unleashed 🌋 @GoogleDeepMind - pushing the scale and dexterity of tasks on our ALOHA 2 fleet. Here is a thread with some of the coolest videos! The first task is hanging a shirt on a hanger (autonomous 1x)
ALOHA 2 is here! 🤖 aloha-2.github.io The upgraded hardware has: * More durability * Greater fine-grained control * Lower stress on operator All off-the-shelf + open-source.
ALOHA 2 is here! 🤖 aloha-2.github.io The upgraded hardware has: * More durability * Greater fine-grained control * Lower stress on operator All off-the-shelf + open-source.
Led by @GoogleDeepMind, we present ALOHA 2 🤙: An Enhanced Low-Cost Hardware for Bimanual Teleoperation. ALOHA 2 🤙 significantly improves the durability of the original ALOHA 🏖️, enabling fleet-scale data collection on more complex tasks. As usual, everything is open-sourced!
What happens when we train the largest vision-language model and add in robot experiences? The result is PaLM-E 🌴🤖, a 562-billion parameter, general-purpose, embodied visual-language generalist - across robotics, vision, and language. Website: palm-e.github.io
Digit Significance in Machine Learning by Vincent Vanhoucke in @TDataScience towardsdatascience.com/digit-signific…
小さなビートメイカー
A summary of our recent work on the use of convex duality (specifically the use of Q-LP with convex regularization) in RL: ai.googleblog.com/2020/07/dualit… via @GoogleAI
This is an awesome explanation of our paper! Especially loved your description of the Temporal Self-similarity Matrix. You end up answering some questions that I was asked during the conference. Thanks for taking the time and effort @ykilcher
This is an awesome explanation of our paper! Especially loved your description of the Temporal Self-similarity Matrix. You end up answering some questions that I was asked during the conference. Thanks for taking the time and effort @ykilcher
Welcome to Alphabet!
Welcome @rob_fergus to @DeepMind. He's a computer vision and deep learning pioneer & a leading scientist in AI. I’m delighted he’s joined @DeepMind (remaining Prof @ NYU), looking forward to working with him again many years after we were at NYU.
Check out RepNet, a single model that analyzes video to provide counting statistics and identify changes in patterns for a broad range of repeating processes—exercising, a bird flapping its wings, pendulums swinging & more. Read more and try it out at goo.gle/2CxtSkn
Introducing RepNet, a model that counts repetitions in videos of *any* action w @yusufaytar, @JonathanTompson, @psermanet and Andrew Zisserman Paper: openaccess.thecvf.com/content_CVPR_2… Project: sites.google.com/repnet Video: youtube.com/watch?v=bvqg-n… #CVPR2020 #computervision #deeplearning
arxiv.org/abs/1912.05032 <-- my most recent paper with @ikostrikov & J. Tompson introducing *ValueDICE* - an off-policy imitation learning alg. We set new SOTA for online imitation learning, and for the 1st time (afaik) beat behavior cloning in the totally offline regime.
@ikostrikov is presenting this at ICLR today! Join our zoom poster session to learn about ValueDICE -- a new & highly-performant off-policy imitation learning algo! 10am-12 & 1-3pm PST.
@ikostrikov is presenting this at ICLR today! Join our zoom poster session to learn about ValueDICE -- a new & highly-performant off-policy imitation learning algo! 10am-12 & 1-3pm PST.
How to scale-up multi-task learning? Self-supervise plan representations from lots of cheap unlabeled play data (no RL was used). learning-from-play.github.io by @coreylynch, Mohi Khansari, @xiao_ted, @Vikashplus, Jonathan Tompson, @svlevine and @psermanet
How to scale-up multi-task learning? Self-supervise plan representations from lots of cheap unlabeled play data (no RL was used). learning-from-play.github.io by @coreylynch, Mohi Khansari, @xiao_ted, @Vikashplus, Jonathan Tompson, @svlevine and @psermanet
Excited to share our work on self-supervised learning in videos. Our method, temporal cycle-consistency (TCC) learning, looks for similarities across videos to learn useful representations.#CVPR2019 #computervision Video: youtube.com/watch?v=iWjjeM… Webpage: sites.google.com/corp/view/temp…
Brian Clark 👹 @troll1184
741 Followers 652 Following I am a software developer, and co-founder of an investment advisory practice and private fund specializing in Vol derivatives.
Masato Kobayashi @MeRTcookingEN
458 Followers 96 Following Assistant Professor |Robotics & AI | Ph.D. 🤖 |Osaka University, Japan | IEEE IES Best Conf. Paper|JP → @MeRTcooking | This account mainly tweets in English.
Martin @visgean
71 Followers 210 Following
Antti Jalomäki @AJalomaki
2K Followers 2K Following Prototyping robots and consulting. Ex-expat and energy. Panicking hype investor. it/its 🌻
Silvio Traversaro @sitraversaro
922 Followers 5K Following Robotics Applied Scientist @ami_iit @iCub @IITalk . Robotics Software @condaforge @RoboStack .
Victor Ng @victor94158
30 Followers 438 Following
Ahmet S Asarkaya @asark38
27 Followers 73 Following
Nithin Varghese @call_me_Nithin
117 Followers 1K Following machine learning/DL practitioner. Believes in a beautiful future supported by A.I. views my own.
Cory Levy @cory
23K Followers 3K Following entrepreneur and early stage investor | Z Fellows @zfellows_
Nalbandian @Joe__Black__
71 Followers 402 Following
Binghao Huang @binghao_huang
667 Followers 945 Following CS PhD student @UIUC, advised by @YunzhuLiYZ. Previous Mechanical and Aerospace Engineering @UCSD, advised by @xiaolonw. Embodied AI/Robot Learning
Jacob Zietek @JacobZietek
113 Followers 321 Following robots, ml, human-robot collaboration | ex-Amazon, Shield AI
Kuter Dinel @KuterDinel
142 Followers 624 Following 2022 SWE Intern at @Google . I like to tweet about things I learn, observe, and think
Oussama Fezzai @OFezzai
54 Followers 319 Following Master’s in machine vision and artifical intelligence || Engineer in automation and industrial computing || Data Analyst
Deping Zhang @joebradly
101 Followers 3K Following
9SManagement @9SManagement
134 Followers 1K Following Treballant per l'eficiència en la Gestió Empresarial
hugohann @hugoohann
0 Followers 2K Following
Tianli Ding @TianliDing
83 Followers 257 Following
Dibyajyoti Acharya @DibyajyotiAch04
136 Followers 5K Following Student, Learner, Explorer 🤓 Interested in all things AI.
lorenzoterenzi @lterenzi96
66 Followers 929 Following
Jacky Liang @jackyliang42
4K Followers 728 Following Research Scientist @GoogleDeepMind working on foundation models for robotics. PhD @CMU_Robotics @iamlab_cmu. Prev. intern @NVIDIAAI. Writes @Last_Week_in_AI
William Lee @William_J_Lee
31 Followers 154 Following
Grace @Graceming04
1K Followers 3K Following I used to be an investment advisor, but I have transformed in a larger environment. After working at Coinbase for 3 years, I became an outstanding analyst. 💗💓
Retoc @retoc71586
17 Followers 106 Following Robotics MSc @ETH Zurich. Figuring out a way of bringing robots into our life.
ADITYA ROY @TrulyAdi
292 Followers 5K Following "Realm" will see the light of day. Reach out : https://t.co/lQA37PppYa
Mr. Jack Tung @MrJackTung
208 Followers 3K Following
Dan Grover @DanGrover
11K Followers 4K Following Product guy, now working on robotics + AI @ @standardbots. Prev. Meta, Tencent.
Tony Z. Zhao @tonyzzhao
12K Followers 765 Following CS PhD student @Stanford and student researcher @GoogleDeepMind. Aspiring full-stack roboticist. Prev Tesla, GoogleX, Berkeley.
Kamyar Ghasemipour @coolboi95
827 Followers 2K Following ML grad student at U of T / Vector Institute. I go by Kamyar. Currently affiliated with Google Robotics. He/Him.
Caleb Martin @Digital_DAT
534 Followers 2K Following Building intelligent robots. Pro Human/humanity 🇺🇸
Tom Primožič 🪐�.. @tomprimozic
357 Followers 1K Following optimist, programmer, mathematician interested in health, longevity, biotech, AI, space, robotics
jack harmer @singularityEvnt
150 Followers 497 Following Co-founder and CTO at Latent Technology. Building virtual robots that understand physics.
Pierre Nagorny @a1rb4ck
415 Followers 5K Following Machine Learning @artanim_mocap #VR #animation #mocap. Happy with real-life optimization problems. @[email protected]
Terminal Fence Sitter @rigiddigid
69 Followers 315 Following i'm a secret third thing all roads lead to BCI
Dipak Khatri @DipakKh41365268
218 Followers 3K Following
jiawei @sk413025
127 Followers 2K Following
👾 @the_dismal_tide
307 Followers 5K Following “Mathematics as currently practiced is a delicate interplay between monastic contemplation and blowing stuff up with dynamite.”
William Lamkin @WilliamLamkin
6K Followers 6K Following 🧙♂️ Navigator of Latent Space 🌌 • ⚗️ Synthesizer of Discourse 📚 • 🗝 Curator of Wisdom 🧠 • 🏕 Keeper of the Flame 🔥
Shenao Zhang @ShenaoZhang
270 Followers 954 Following PhD student @NorthwesternU | Student Researcher @MSFTResearch. Ex-intern @MSFTReserch, ByteDance, and Tencent AI | Previously @GeorgiaTech. LLM, RL, agent.
Matin Moezzi @MatinMoezzi
17 Followers 244 Following Reinforcement Learning Researcher @ University of Toronto
Gagana B @Gagana95086971
33 Followers 2K Following I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant! (Alan Greenspan said it first!)
Pete Florence @peteflorence
3K Followers 361 Following research scientist @GoogleDeepMind // robotics + AI // PhD @MIT_CSAIL
Simon Holland @simoncholland
171K Followers 1K Following My wife bought low sodium bacon and now I know what it means when someone says they love you but aren’t IN LOVE with you.
Ofir Nachum @ofirnachum
4K Followers 342 Following Research at @OpenAI. Previously at @GoogleAI on the Brain Team. Doing work on #ReinforcementLearning and #MachineLearning
Raacked @raacked
43 Followers 179 Following An appreciation page to racked rides. We know that the crown jewel of car mods are roof racks. Its what distinguishes the car, from a car. #raacked
Tom Sercu @TomSercu
2K Followers 693 Following building something new and hiring. Ex-Meta FAIR, Ex-IBM Research. Alum @NYU, @ugent.
Ted Xiao @xiao_ted
11K Followers 679 Following I teach robots to be smarter @GoogleDeepMind. Tweets about robot learning, scaling, and large models. Opinions my own.
joan bruna @joanbruna
3K Followers 231 Following Associate Professor of Computer Science, Data Science & Mathematics (aff), Courant Institute and Center for Data Science, NYU. He/His/Him
Kevin Patrick Murphy @sirbayes
42K Followers 328 Following Research Scientist at Google Brain / Deepmind. Interested in Bayesian Machine Learning.
AI at Meta @AIatMeta
523K Followers 254 Following Together with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.
Chelsea Finn @chelseabfinn
69K Followers 384 Following Asst Prof of CS & EE @Stanford. PhD from @Berkeley_EECS, EECS BS from @MIT
Karol Hausman @hausman_k
22K Followers 140 Following @Physical_int ex: researcher @GoogleAI/@DeepMind, adj. Prof. @Stanford. Into robots, AI, NBA, philosophy, soccer and almond croissants. 🇵🇱🇺🇸
Jason Yosinski @jasonyo
7K Followers 912 Following Co-founder/CEO @Windscape_AI, Co-founder/Pres @ml_collective, investor, adviser. Prev: Uber AI Labs founding team, Cornell, Montreal, Caltech 🌻
Andy Zeng @andyzeng_
6K Followers 443 Following Building smarter robots @GoogleDeepMind. PhD @Princeton. CS & Math @UCBerkeley
koray kavukcuoglu @koraykv
8K Followers 84 Following VP of Research and Technology at Google DeepMind
Yann LeCun @ylecun
706K Followers 714 Following Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
Clément @clmt
15K Followers 2K Following AI @ Google DeepMind. Ex NVIDIA (built AI for self-driving cars + GPU data science), Twitter (started AI team), MadBits (founded+sold @ Twitter) 🇺🇸🇫🇷
Ilya Kostrikov @ikostrikov
8K Followers 615 Following Researcher @OpenAI, previously @Postdoc at UC Berkeley @berkeley_ai, PhD in CS @CILVRatNYU
Corey Lynch @coreylynch
10K Followers 1K Following AI at @figure_robot, previously research scientist at @GoogleDeepMind.
Laura Graesser @lgraesser3
949 Followers 525 Following Learning about robots learning. Researcher, Robotics at Google DeepMind. Co-author of Foundations of Deep Reinforcement Learning. https://t.co/JgfxI9VJAh
Debidatta Dwibedi @debidatta
845 Followers 2K Following Senior Research Scientist @GoogleDeepMind, Previously Robotics @CarnegieMellon, EE @IITKanpur, StudApps (https://t.co/iDVr86IjhA)
Eric Jang @ericjang11
68K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0pIntroducing 𝐀𝐋𝐎𝐇𝐀 𝐔𝐧𝐥𝐞𝐚𝐬𝐡𝐞𝐝 🌋 - Pushing the boundaries of dexterity with low-cost robots and AI. @GoogleDeepMind Finally got to share some videos after a few months. Robots are fully autonomous filmed in one continuous shot. Enjoy!
For the past year we've been working on ALOHA Unleashed 🌋 @GoogleDeepMind - pushing the scale and dexterity of tasks on our ALOHA 2 fleet. Here is a thread with some of the coolest videos! The first task is hanging a shirt on a hanger (autonomous 1x)
ALOHA 2 is here! 🤖 aloha-2.github.io The upgraded hardware has: * More durability * Greater fine-grained control * Lower stress on operator All off-the-shelf + open-source.
Led by @GoogleDeepMind, we present ALOHA 2 🤙: An Enhanced Low-Cost Hardware for Bimanual Teleoperation. ALOHA 2 🤙 significantly improves the durability of the original ALOHA 🏖️, enabling fleet-scale data collection on more complex tasks. As usual, everything is open-sourced!
Led by @GoogleDeepMind, we present ALOHA 2 🤙: An Enhanced Low-Cost Hardware for Bimanual Teleoperation. ALOHA 2 🤙 significantly improves the durability of the original ALOHA 🏖️, enabling fleet-scale data collection on more complex tasks. As usual, everything is open-sourced!
What happens when we train the largest vision-language model and add in robot experiences? The result is PaLM-E 🌴🤖, a 562-billion parameter, general-purpose, embodied visual-language generalist - across robotics, vision, and language. Website: palm-e.github.io
I'm thrilled to announce I have joined @DeepMind to help their extraordinary team continue to lead in their quest to build AGI. More to come. I'm also so happy to reconnect with so many friends I worked with a decade ago. Let the adventure begin!
Don't let low-order statistics fool you
Wondering why the first layer of some recent DL architectures for vision are called "linear embedding of 16x16 non-overlapping patches" instead of "Convolutional layer with 16x16 kernels and 16x16 stride" ???
Quantifying uncertainty is critical for good offline RL. For model-based methods, this means representing the uncertainty of the model’s predictions. This is unfortunately quite hard when you move to image observations! (2/4)
Finally, we took an offline dataset from a previous paper by @anniee268, Nam, @SurajNair_1 et al., labeled 200 images with sparse rewards, and trained the robot with LOMPO to close the drawer of a desk *without any further data*. The resulting policy works quite well. 🤖👇 (4/4)
Interested in offline RL? Handling image observations & continuous actions is important for offline RL in the real world. We introduce LOMPO to tackle this setting. arxiv.org/abs/2012.11547 with Rafael Rafailov, @TianheYu, @aravindr93 🧵👇(1/4)
if you can't handle me at iron 1 you don't deserve me at radiant
Butterfly's VP of AI @NCSilberman presents a keynote at #medneurips @NeurIPSConf tomorrow at 8am. Tune in to learn about real-world insights from developing machine learning models for ultrasound. sites.google.com/view/med-neuri…
Thanks to @chelseabfinn for the great conversation about her work on meta-learning and robotics -- check it out below!
Episode 10 of The Thesis Review: Chelsea Finn (@chelseabfinn), "Learning to Learn with Gradients" We discuss meta-learning, her work on MAML and its applications, and the future of robotics research soundcloud.com/thesis-review/…
@ArashVahdat @JyotiAneja @alexschwing @jankautz awesome work! you should checkout energy-inspired models (arxiv.org/abs/1910.14265), which gives a way to compute likelihood for these distributions. wonder if you get bpd improvements too? p.s. the term noise-contrastive priors was used here too: arxiv.org/abs/1807.09289 🙃
I'm on O-1 visa myself so probably a biased opinion here, but I really don't understand how you can justify this policy.
I've personally known many people in the US on O-1 visas & worked closely with many of them. I can say very confidently that holders of this particular kind of visa truly are people of *extraordinary* ability and we want these scientists, artists, and other skilled people here!
Me trying to reason about Machine Learning without knowing anything about real analysis or measure theory
What's the name for that feeling when you have an idea and think it may work but don't actually want to check it for fear it'll just vanish or prove bogus?
1) strong accept 2) accept 3) reject 4) strong reject 5) classical approach
There is no greater diss than your work being referred to in a deep learning paper as a "classical approach"
Trends for United States
You might like
