Ashvin Nair @ashvinair
Research scientist @openai. Prev: 9 years learning to poke by poking at UC Berkeley ashvin.me Berkeley, CA Joined December 2014-
Tweets64
-
Followers760
-
Following2K
-
Likes291
OpenAI is nothing without its people
congrats @ashvinair ! proud of you and the team behind code interpreter. selfishly now I can finally be a DS expert without taking a stats class 🙃 nytimes.com/2023/07/11/tec…
Super excited to share #CodeInterpreter with all #ChatGPT plus users! This allows chatgpt to take actions on your behalf (via the universal action space of code) with your data.
Code Interpreter will be available to all ChatGPT Plus users over the next week. It lets ChatGPT run code, optionally with access to files you've uploaded. You can ask ChatGPT to analyze data, create charts, edit files, perform math, etc. Plus users can opt in via settings.
Super excited for ChatGPT plugins - a way for LLMs to interact with the external world
Super excited for ChatGPT plugins - a way for LLMs to interact with the external world
Offline RL initialization on diverse data can make online robotic RL far more practical! In our new paper, we show that this works great for industrial connector insertion, pretraining on dozens of connectors and finetuning autonomously to a new one! A thread:
We will present Planning to Practice (PTP) at IROS 2022 this week. Check out our paper if you're interested in using visual affordances learned from prior data to facilitate policy learning for novel sequential manipulation tasks.
We will present Planning to Practice (PTP) at IROS 2022 this week. Check out our paper if you're interested in using visual affordances learned from prior data to facilitate policy learning for novel sequential manipulation tasks.
Pretraining on diverse robot data with offline RL allows planning multi-stage tasks, and then finetuning those tasks with end-to-end exploration. Our new paper, FLAP, describes a method for doing this with real robots! arxiv.org/abs/2210.06601 sites.google.com/view/project-f… Thread:
Planning to practice (PTP) combines goal-based RL with high-level planning over image subgoals to finetune robot skills. Goal-conditioned RL provides low-level skills, and they can be integrated into high-level skills by planning over "affordances": arxiv.org/abs/2205.08129 🧵->
ArXiv arxiv.org/abs/2205.08129: Training goal-conditioned policies that reach goals on command. 1) Decompose problem by a planner that suggests subgoals. 2) Hybrid approach pretrains subgoal generator and policy by offline reinforcement learning, then fine-tunes policy online.
Representation learning via goal-conditioned bisim can enable "analogies" between different scenes. In our new work, we explore how to learn representations that provide analogies, then apply it to goal-conditioned RL. See @yayitsamyzhang's thread: Paper: arxiv.org/abs/2204.13060
Representation learning via goal-conditioned bisim can enable "analogies" between different scenes. In our new work, we explore how to learn representations that provide analogies, then apply it to goal-conditioned RL. See @yayitsamyzhang's thread: Paper: arxiv.org/abs/2204.13060
Building generalizable goal-conditioned agents from rich observations is a key to solving real world problems. We propose a new form of state abstraction, goal-conditioned bisimulation, that captures functional equivariance. 1/n Paper: arxiv.org/abs/2204.13060
Join our #ICLR2022 workshop on Generalizable Policy Learning in the Physical World this Friday, Apr 29th @ 8:00 AM PDT (UTC -7) ! ✅ Web: ai-workshops.github.io/generalizable-… 🙋🏻♀️ Submit questions: forms.gle/95TVhPeR2pop7m…
Consider submitting to this ICLR workshop on generalizable policy learning!
Consider submitting to this ICLR workshop on generalizable policy learning!
Implicit Q-learning, or "I can't believe it's not SARSA": state-of-the-art offline RL results, fast and easy to implement; almost SARSA, but with a different loss to provide "implicit policy improvement": arxiv.org/abs/2110.06169 w/ @ikostrikov, @ashvinair 🧵->
Excited to present our work with @ashvinair and @svlevine, Offline RL with Implicit Q-Learning (IQL), a simple method that achieves SOTA performance on D4RL arxiv.org/abs/2110.06169 and works 4x faster than prior SOTA github.com/ikostrikov/imp… Thread below
Really excited about new work with @ikostrikov and @svlevine on offline RL. Implicit Q-Learning (IQL) gets great results offline and finetuning further online, and is really fast and simple to implement. See Ilya's thread for details
Really excited about new work with @ikostrikov and @svlevine on offline RL. Implicit Q-Learning (IQL) gets great results offline and finetuning further online, and is really fast and simple to implement. See Ilya's thread for details
[1] @AlexKhazatsky , @ashvinair and @svlevine on learning skills by imagining visual affordances: bair.berkeley.edu/blog/2021/09/2…
How do we get robots to learn what they *can* do from prior data, so that they can learn what they *should* do via RL? Visual affordance learning (VAL) does this via goal-conditioned RL and generative models. New blog post by @ashvinair & @AlexKhazatsky bair.berkeley.edu/blog/2021/09/2…
Jim Fan @DrJimFan
229K Followers 3K Following @NVIDIA Sr. Research Manager & Lead of Embodied AI (GEAR Lab). Creating foundation models for Humanoid Robots & Gaming. @Stanford Ph.D. @OpenAI's first intern.
Sergey Levine @svlevine
79K Followers 122 Following Associate Professor at UC Berkeley Co-founder, Physical Intelligence
Eugene Vinitsky @EugeneVinitsky
13K Followers 2K Following Anti-cynic. Artificial narrow intelligence. Autonomous vehicles, multi-agent learning, and transportation. RS at Apple, Asst. Prof at @nyutandon. He/him.
Ted Xiao @xiao_ted
11K Followers 681 Following I teach robots to be smarter @GoogleDeepMind. Tweets about robot learning, scaling, and large models. Opinions my own.
Natasha Jaques @natashajaques
25K Followers 1K Following Senior Research Scientist at @GoogleAI and Assistant Professor @uwcse. Social Reinforcement Learning in multi-agent and human-AI interactions. PhD from @MIT.
Abhishek Gupta @abhishekunique7
5K Followers 639 Following Assistant Professor at University of Washington. I like robots, and reinforcement learning. Previously: post-doc at MIT, PhD at Berkeley
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
Nathan Lambert @natolambert
25K Followers 689 Following Figuring out AI @allen_ai, "rl boi" DM me papers. Writes @interconnectsai, talks @retortai Has phd and some credentials
Xiaolong Wang @xiaolonw
11K Followers 947 Following Assistant Professor @UCSDJacobs Postdoc @berkeley_ai PhD @CMU_Robotics
Michael Chang @mmmbchang
2K Followers 2K Following Amplify human creativity Gemini @GoogleDeepMind Prev @LangChainAI, @MetaAI, @SchmidhuberAI PhD @berkeley_ai, w/ @svlevine, @cocosci_lab. BS @MIT @MITCoCoSci
Jacky Liang @jackyliang42
4K Followers 733 Following Research Scientist @GoogleDeepMind working on foundation models for robotics. PhD @CMU_Robotics @iamlab_cmu. Prev. intern @NVIDIAAI. Writes @Last_Week_in_AI
Ken Goldberg @Ken_Goldberg
16K Followers 8K Following artist, @UCBerkeley engineering prof, @AmbiRobotics chief scientist, robots, rockets, redwoods, rebels.
Vikash Kumar ✈️IC.. @Vikashplus
4K Followers 406 Following Studying intelligent embodied behaviors. Ad. Prof. @CMU_Robotic | Sr. research scientist at @AIatMeta @GoogleAI @OpenAI | @berkeley_ai @UWcse #MuJoCo
Joshua Achiam ⚗️ @jachiam0
14K Followers 945 Following Human. Trying to make safe alchemy machines. Thinking about humanist alchemism (h/alc ⚗️, maybe). Main author of https://t.co/cKuSh210l1
Markus Wulfmeier @m_wulfmeier
12K Followers 1K Following machine learning & the real world @GoogleDeepMind - european @ELLISforEurope - transfer and generalisation - priors: @oxfordrobots @berkeley_ai @ETH @MIT
Keerthana Gopalakrish.. @keerthanpg
13K Followers 819 Following Building Embodied AGI. Research @DeepMind. Opinions my own.
Dinesh Jayaraman @dineshjayaraman
2K Followers 571 Following Assistant Professor at University of Pennsylvania. Computer vision, machine learning, robotics.
Tisusays @tisusays13070
0 Followers 177 Following
Anurag Ajay @aajay3110
90 Followers 109 Following PhD student at MIT. Interested in Generative Modeling and Multi-Modal reasoning. Interned at MetaAI (FAIR) and Google Brain.
Veola Schwartze @VeolSchwar
69 Followers 5K Following
Raghu Mocharla @RaghuMocharla
84 Followers 284 Following
Rania Dixon @RaniDixon
39 Followers 5K Following
Admirable @DeniseCole7912
2 Followers 304 Following
Erika Shingleton @ShingletoErik
29 Followers 5K Following
kolergy @kolergy
211 Followers 2K Following Test early, physics always win at the end... Machine learning is changing the world! Be prepared
fatweir_00do @fatweir1869
20 Followers 1K Following
Eleni Redenz @EleniReden
53 Followers 5K Following
Krishna Suresh @_krishnasuresh
24 Followers 257 Following
dipsu kilme @DipsuKilme
102 Followers 245 Following
Mevan Solanga @MevSolanga
1 Followers 20 Following
Kenny Shaw @kenny__shaw
507 Followers 640 Following 1st Year PhD (fall 2023) in Robotics at Carnegie Mellon with Prof. Deepak Pathak. Interested in Robot Hands. NSF GRF.
Jacob Zietek @JacobZietek
132 Followers 390 Following AI Engineer @armada_ai | robots, ai, human-robot collaboration | ex-Amazon, Shield AI | 🇺🇸
Ali K @alihkw_
263 Followers 1K Following DMs Open! AI & Robotics MSc at @Mila_Quebec/@UMontreal. Prev: nlp at @amazonscience, compsci at @UofT. Slowly figuring out how to make robots learn.
Izabela Lindersmith @ILindersmi25871
54 Followers 5K Following
Elsie-may Buswell @m_buswe
81 Followers 5K Following
Chasidy Strem @ChasidyStr6513
86 Followers 5K Following
Ashwin Ramaswami @ashwinforga
3K Followers 2K Following Candidate for GA State Senate District 48. Contact me at [email protected]
us_Hope_ @usHope114619
6 Followers 969 Following
Gunbir Singh Baveja @g_baveja
16 Followers 85 Following visiting researcher @kaist_ai advised by @JosephLim_AI; sophomore @UBC
Alexa Mclaen @AlexaM64811
45 Followers 5K Following
Eliza_DFG @ElizaD42357
5 Followers 841 Following
Priyanka_21 @Priyank18753398
2 Followers 753 Following
Kris856 @Kris856290899
39 Followers 933 Following
sandypadaiyachi993 @sandypadai49724
12 Followers 967 Following
Dr_krishna68413 @krishna68467178
1 Followers 1K Following
RadiantSway @RadiantSwa61747
4 Followers 740 Following
Shirleen Dubitsky @SDubitsky13106
39 Followers 5K Following
Emma Charm @charm_emma70594
2 Followers 761 Following
Fearless_Femme @FearlessFe72605
30 Followers 1K Following
_Petrichor @Petrichor436343
6 Followers 744 Following
Amira Doucette @DoucettAmi
47 Followers 5K Following
Shiva @ShivaSujit
387 Followers 609 Following Deep RL at @ArayaGlobal | Prev Intern @MSFTResearch | MSc @Mila_Quebec in RL | BSc @ReachNITT
Dawn_SDF @SdfDawn80252
9 Followers 1K Following
Cecelia Cobham @CobhamCobh
55 Followers 5K Following
Juan Hmmm @JuanAH03488233
76 Followers 3K Following
Mohammed Mudassir @mohd_mudassir71
194 Followers 4K Following Full Stack Developer | Java | JavaScript | Node.js | React.js | Express | Senior Software Engineer @Oracle | Ex-@dassaultSystemes
us_Kara_ @UsKara85299
2 Followers 1K Following
Judith Kruss @judit_kru
17 Followers 2K Following
Susana Westerholm @SusaWesterhol
73 Followers 5K Following
GreenGoddess_ @GreenG35851
5 Followers 465 Following
Jim Fan @DrJimFan
229K Followers 3K Following @NVIDIA Sr. Research Manager & Lead of Embodied AI (GEAR Lab). Creating foundation models for Humanoid Robots & Gaming. @Stanford Ph.D. @OpenAI's first intern.
Yann LeCun @ylecun
710K Followers 718 Following Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
Karol Hausman @hausman_k
22K Followers 141 Following @Physical_int ex: researcher @GoogleAI/@DeepMind, adj. Prof. @Stanford. Into robots, AI, NBA, philosophy, soccer and almond croissants. 🇵🇱🇺🇸
Danijar Hafner @danijarh
14K Followers 869 Following Building AI that makes autonomous decisions using world models, artificial curiosity, and temporal abstraction @DeepMind
Animesh Garg @animesh_garg
21K Followers 1K Following Foundation Models for Generalizable Autonomy. Assistant Professor in AI Robotics @GeorgiaTech + @NvidiaAI. prev @Stanford @berkeley_ai @UofTCompSci
Andrej Karpathy @karpathy
977K Followers 904 Following 🧑🍳. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets 🧠🤖💥
Sergey Levine @svlevine
79K Followers 122 Following Associate Professor at UC Berkeley Co-founder, Physical Intelligence
Google DeepMind @GoogleDeepMind
942K Followers 275 Following We’re a team of scientists, engineers, ethicists and more, committed to solving intelligence, to advance science and benefit humanity.
Eugene Vinitsky @EugeneVinitsky
13K Followers 2K Following Anti-cynic. Artificial narrow intelligence. Autonomous vehicles, multi-agent learning, and transportation. RS at Apple, Asst. Prof at @nyutandon. He/him.
Conference on Robot L.. @corl_conf
7K Followers 38 Following Conference on Robot Learning (CoRL) is an annual international conference on robotics and machine learning. Account managed by @gmartius
Ted Xiao @xiao_ted
11K Followers 681 Following I teach robots to be smarter @GoogleDeepMind. Tweets about robot learning, scaling, and large models. Opinions my own.
Natasha Jaques @natashajaques
25K Followers 1K Following Senior Research Scientist at @GoogleAI and Assistant Professor @uwcse. Social Reinforcement Learning in multi-agent and human-AI interactions. PhD from @MIT.
Abhishek Gupta @abhishekunique7
5K Followers 639 Following Assistant Professor at University of Washington. I like robots, and reinforcement learning. Previously: post-doc at MIT, PhD at Berkeley
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
Chris Paxton @chris_j_paxton
8K Followers 1K Following Mostly posting about robots. Embodied AI @hellorobotinc, formerly @AIatMeta, @NVIDIAAI, @zoox. All views my own.
Paul Graham @paulg
1.9M Followers 772 Following
Trenton Bricken @TrentonBricken
6K Followers 2K Following Trying to figure out what makes minds and machines go "Beep Bop!" @AnthropicAI
Sophia Sanborn @naturecomputes
4K Followers 3K Following Theory, ML, neurotechnology @ https://t.co/OmhC0RyxZp | Organizer @neur_reps | Prev: @geometric_intel @berkeley_ai @redwood_neuro @intelai @harvard
Keerthika Uthayakumar @keerthikau
7K Followers 790 Following nba stats, raptors | AD @tsn_sports | Not all fun facts are fun, some are red flags | email: [email protected]
leehsienloong @leehsienloong
890K Followers 29 Following Prime Minister of Singapore and leader of the @papsingapore. This account is maintained by the Prime Minister’s Office. Tweets by Mr Lee are signed “LHL”.
Nicole Shanahan @NicoleShanahan
38K Followers 3 Following Mom to Echo; Healthy Planet & Healthy Humans
Jon Denton @JonDenton
56K Followers 1K Following YouTuber. Cynical optimist. Over 470k subs. Chairman of keeping it real 25/8. https://t.co/PRb2V9pUEu
Pessimists Archive @PessimistsArc
91K Followers 64 Following Exploring technophobia and moral panic through the ages. A litany of shameful cynicism and spite. Curated by @louisanslow
RainnWilson @rainnwilson
4.3M Followers 674 Following Just another luminous spiritual being having a beautiful and difficult human experience.
Meaning Alignment Ins.. @meaningaligned
931 Followers 10 Following The Meaning Alignment Institute is a research organization with the goal of ensuring human flourishing in the age of AGI.
FAR AI @farairesearch
1K Followers 19 Following Ensuring AI systems are trustworthy and beneficial to society by incubating new AI safety research agendas.
Daniel Faggella @danfaggella
7K Followers 238 Following Host of The Trajectory: Realpolitik on AGI and the posthuman transition (https://t.co/1rCgmOZ2EA). Founder @Emerj AI Research.
cyriak harris @cyriakharris
66K Followers 272 Following animator of weird stuff. https://t.co/mpRSimd5Oa https://t.co/7zxr7CxxmZ
Mouin Rabbani @MouinRabbani
56K Followers 483 Following
Physical Intelligence @physical_int
4K Followers 8 Following Physical Intelligence (Pi), bringing AI into the physical world.
Rocky Duan @rocky_duan
782 Followers 84 Following Building @CovariantAI, CTO. Previously @OpenAI, @UCBerkeley PhD. 2024 Forbes 30 Under 30.
Yishan @yishan
79K Followers 355 Following I run Terraformation, and I was once the CEO of Reddit. Both are very interesting challenges. Views are mine alone, but also yours if I do my job right.
Lisa @LisaShmulyan
996 Followers 1K Following Chief of Staff @everybanking | Also building SF's community for the most ambitious mission driven founders under 26 @sigma2society
Groq Inc @GroqInc
44K Followers 467 Following Creator of the LPU™ Inference Engine, providing the fastest speed for AI applications, designed & engineered in N. America https://t.co/DsEqVAC5Dp
Karl Pertsch @KarlPertsch
1K Followers 206 Following Robot Foundation Models @ UC Berkeley & Stanford | Postdoc w/ Sergey Levine & Chelsea Finn | Prev. Intern @ Google Brain, Meta AI | PhD @ USC.
Chuer Pan @chuer_pan
561 Followers 609 Following Robotics PhD @Stanford | MSR @CMU_Robotics | @VectorInst, EngSci Robo Grad @UofT
Romain Huet @romainhuet
21K Followers 7K Following Head of Developer Experience @OpenAI. Previously, Product Lead @Stripe, Platform @Twitter, Co-Founder & CTO of Jolicloud.
MyoSuite @MyoSuite
570 Followers 34 Following In pursuit of human level dexterity & agility in physiological human digital twins
Qinqing Zheng @qqyuzu
226 Followers 110 Following Reinforcement Learning, Generative Modeling @ FAIR (@AIatMeta). PhD @UChicago.
lele @CherrilynnZ
131 Followers 273 Following if an exclamation point was a person and then put under immense duress / design and psychology @cal
Will Stancil @whstancil
81K Followers 508 Following I research metro policy and civil rights, focused on housing and schools. Proud member of Do-Something Twitter. Running for MN House! https://t.co/Q5xzs7kUmw
fry @anniefryman
10K Followers 2K Following “You live in a society!” 🌎 designer → policy person, now focused on sf housing and transportation @spur_urbanist 📐
OpenAI Developers @OpenAIDevs
71K Followers 0 Following Official @OpenAI account for anyone building on our APIs. Join us in building the future of AI. We ❤️ developers!
Vishal Kuo @vishalkuo
652 Followers 1 Following
Milei Explains @Milei_Explains
45K Followers 314 Following Milei in English. Translating Milei videos for friends. No politics. 100% Economics. Not affiliated with Javier Milei. All clips fair use.
valencia4people @valencia4people
704 Followers 175 Following From Market to Cesar Chavez, bursting with @SharedSpacesSF flavor & fun, longing to shed motor vehicles and become “Valencia For The People” 📷 @irapolis
Max Kleiman-Weiner @maxhkw
4K Followers 740 Following computational cognitive scientist working on social and artificial intelligence. professor @UW, cofounder @CSM_ai, PhD @MIT
parm @prmshra
10K Followers 2K Following breaking a paradigm 🦦 . epi-transcriptomics 🧬 . student. researcher. @penn computational biology, ML. 🇮🇳 🇺🇸
East Bay Transit Ride.. @TransitRidersEB
949 Followers 42 Following East Bay Transit Riders Union! Building rider-power on AC Transit! Supporting riders and employees! Public streets 4 Public transit first.
Safe Street Rebel @SafeStreetRebel
8K Followers 837 Following End car dominance and save the planet one direct action at a time. Likes: bus lanes, bike routes. Dislikes: cops, the petrostate @[email protected]
Kanika Dutta @Kanika_Dutta_
4K Followers 933 Following Product @noviconnect • Scout @aforevc • Engineering @UWaterloo
NBEinstein @NBEinstein
26K Followers 763 Following NBA - Video Breakdown (French version of me here : @GuillaumeBLab) 🔴Youtube : https://t.co/J0IhCa94ve .
@ashvinair @GoogleDeepMind thanks Ashvin! when are you making your comeback to robotics? haha
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)
Happy Tamil New Year and Happy Vishu! இனிய புத்தாண்டு நல் வாழ்த்துக்கள்💐💐
Sora is our first video generation model - it can create HD videos up to 1 min long. AGI will be able to simulate the physical world, and Sora is a key step in that direction. thrilled to have worked on this with @billpeeb at @OpenAI for the past year openai.com/sora
Today was my last day at @OpenAI. 20 months ago, I left the YC startup I founded to join the Codex team where I prototyped the first code interpreter demos. From there, I helped start the internal Human Data team, where we built tools to produce data for frontier efforts, one…
sent her the kanye leak and she still didnt text me back 😔✊
Proud to share Octo: our fully open robot "foundation" policy trained on >25 M actions across > 20 robots. Octo predicts raw robot actions from observation (from visual → **embodied** pre-training) -- and finetunes fast to wholly new robot setups. octo-models.github.io 🧵
special thanks to @steph_palazzolo @theamberyang @berber_jin1 @surbhirathoree @ashvinair and the @KindoAI team for braving the cold and joining the speaker party! 🚀
Excited to announce our new work on using synthetic data for improving mathematical problem solving and code generation in LLMs! arxiv: arxiv.org/abs/2312.06585 A small amount of fine-tuning can lead to large gains (>6% on Hendrycks MATH with Palm-2)
we have trained a humanoid transformer with large-scale reinforcement learning in simulation and deployed it to the real world zero-shot
2023 is Possibly the greatest year in over 70 years for passenger rail in America 🧵
Just gonna shill my stuff amidst the chaos and speculation: sea-snell.github.io/ILQL_site/
@ilyasut People make mistakes. You will stay my friend.
I deeply regret my participation in the board's actions. I never intended to harm OpenAI. I love everything we've built together and I will do everything I can to reunite the company.
wrote a new essay on what might AI regulation look like? special thanks to @ashvinair @jbenromero open.substack.com/pub/margaretje…
we're so back 🛜
ChatGPT can now browse the internet to provide you with current and authoritative information, complete with direct links to sources. It is no longer limited to data before September 2021.
@tolstoybb Seems decidedly non-communist to think that landowners should decide how their land can be used instead of like 40,000-person-sized blocs collectively deciding how much lawn you're required to have through 15% turnout elections
Trends for United States
You might like
