Faculty Fellow at NYU. Trustworthy (safe x transparent x fair) ML.
I'm on the faculty job market.
Prev: PhD Student & Rhodes Scholar at University of Oxford.timrudner.com NYC + New HavenJoined December 2015
CDS’ @timrudner & Julia Stoyanovich (@stoyanoj) join the national effort to set standards for AI safety with the U.S. AI Safety Institute Consortium.
“The stakes are high. And it’s important to get this right,” said Rudner.
Learn more: nyudatascience.medium.com/cds-tim-g-j-ru…
Just dropped a 4 hour lecture on "Large Language Models": youtu.be/2yjzZfDQxy8
0:00 Basics of language models
2:30 Word2vec
16:27 Transfer Learning
19:23 BERT
1:00:39 T5
1:31:14 GPT1-3
1:53:05 ChatGPT
2:20:03 LLMs as Deep RL
2:53:00 Policy Gradient
3:32:50 Train your…
Seems like everyone wants to know more about how large language models work, but existing explainers often:
-Are too shallow or too technical
-Focus on how LLMs predict the next word, which is only part of the story
So we wrote 3 explainers of our own! 🧵
cset.georgetown.edu/article/the-su…
Making a good benchmark may seem easy---just collect a dataset---but it requires getting multiple high-level design choices right.
@Thomas_Woodside and I wrote a post on how to design good ML benchmarks:
safe.ai/blog/devising-…
📢 We obtained the **first non-vacuous generalization bounds** for pre-trained LLMs! 🚀🤖
Check out our paper (📄: arxiv.org/abs/2312.17173) and @micahgoldblum's excellent thread 🧵below:
📢 We obtained the **first non-vacuous generalization bounds** for pre-trained LLMs! 🚀🤖
Check out our paper (📄: arxiv.org/abs/2312.17173) and @micahgoldblum's excellent thread 🧵below: https://t.co/naaKzCE1lZ
📢 We obtained the **first non-vacuous generalization bounds** for pre-trained large language models! 🚀🤖
Check out our paper (arxiv.org/abs/2312.17173) and @LotfiSanae's excellent thread 🧵 below:
📢 We obtained the **first non-vacuous generalization bounds** for pre-trained large language models! 🚀🤖
Check out our paper (arxiv.org/abs/2312.17173) and @LotfiSanae's excellent thread 🧵 below:
🚨 Model-based methods for offline RL aren’t working for the reasons you think! 🚨
In our new work, led by @anyaasims, we uncover a hidden “edge-of-reach” pathology which we show is the actual reason why offline MBRL methods work or fail!
Let's dive in! 🧵
[1/N]
Please apply if you are a UK undergraduate from under-represented backgrounds interested in exploring what a career in AI research is like. Deadline Feb 17!
@GoogleDeepMind has kindly supported the AI internship projects @uniqplusoxford, thank you!
Please apply if you are a UK undergraduate from under-represented backgrounds interested in exploring what a career in AI research is like. Deadline Feb 17!
@GoogleDeepMind has kindly supported the AI internship projects @uniqplusoxford, thank you!
To help make models more robust and defend against misuse, we created HarmBench, an evaluation framework for automated red teaming and testing the adversarial robustness of LLMs and multimodal models.
🌐 harmbench.org
📝 arxiv.org/abs/2402.04249
[#eacl2024 paper]
TL;DR We introduce 𝗴𝗿𝗮𝗱𝗶𝗲𝗻𝘁-𝗯𝗮𝘀𝗲𝗱 𝗿𝗲𝗱 𝘁𝗲𝗮𝗺𝗶𝗻𝗴 (𝗚𝗕𝗥𝗧), an effective method for triggering language models to produce unsafe responses, even when the LM is finetuned to be safe through 𝑎𝑙𝑖𝑔𝑛𝑚𝑒𝑛𝑡.
MLHC 2024 Cfp is out! Submission deadline is April 9, 2024. We're switching to OpenReview this cycle and the submission site will open soon. Stay tuned!
MLHC 2024 Cfp is out! Submission deadline is April 9, 2024. We're switching to OpenReview this cycle and the submission site will open soon. Stay tuned!
State Space Models: A Modern Approach
This is an interactive textbook on state space models (SSM) using the JAX Python library.
probml.github.io/ssm-book/root.…
"User-friendly Introduction to PAC-Bayes bounds" is now published: nowpublishers.com/article/Detail…
Thanks a lot for all the feedback on the first version I posted on arXiv some time ago 🙏
I'll be teaching a course on this topic at SSML 2024 @OISTedugroups.oist.jp/mlss
I'm hiring!
I'm building 4 research groups under me at AISI (formerly the UK's Taskforce on Frontier AI) to work on foundational AI safety research.
[1/5]
gov.uk/government/pub…
We have now completed our analysis of new materials reported in the Google Deepmind / Berkeley autonomous lab paper. My own initial analysis is in the quote tweet.
Happy to have worked with @SchoopLab to jointly put together a comprehensive analysis, now available on @ChemRxiv.…
We have now completed our analysis of new materials reported in the Google Deepmind / Berkeley autonomous lab paper. My own initial analysis is in the quote tweet.
Happy to have worked with @SchoopLab to jointly put together a comprehensive analysis, now available on @ChemRxiv.…
45K Followers 2K FollowingResearch Director, @VectorInst. Canada CIFAR AI Chair. Associate Professor of Stats/CS @UofT. I study machine learning and AI, emphasis on theory.
689 Followers 335 FollowingPhD student @OxfordStats / @OXCSML supervised by @yeewhye and @wellingmax.
Probabilistic ML, geometric ML and their interestion.
Interned @DeepMind @Qualcomm
60K Followers 2K Followinga combination of a mediocre scientist, a mediocre manager, a mediocre advisor & a mediocre PC at @nyuniversity (@CILVRatNYU) & @genentech (@PrescientDesign).
16K Followers 3K FollowingDirector Data Science Institute @UWMadison @datascience_uw. EiC @MLSTjournal.
Physics, stats/ML/AI, open science.
same handle @sigmoid.social and bsky
33K Followers 1K FollowingDad, spouse, Professor of Machine Learning @UniofOxford, Co-Founder Mind Foundry, Director @aims_oxford. Bayes, Long Covid, porridge, AI must be good for humans
29K Followers 1K FollowingOpen-Endedness Team Lead @GoogleDeepMind, Professor of AI @AI_UCL, PI @UCL_DARK, @ELLISforEurope Scholar. ex @MetaAI (FAIR), @CompSciOxford. Opinions my own.
38K Followers 222 FollowingAssociate Professor of Machine Learning, University of Oxford
@OATML_Oxford Group Leader
Director of Research at AISI (formerly UK Taskforce on Frontier AI)
364 Followers 2K FollowingScientist, physician, necromancer, cajá juice sommelier and real life Shikamaru. I hate 99,9% of people and love all dogs #firstgen
🇧🇷🇨🇦🇪🇦🇸🇪🇳🇿🇰🇷🇿🇦
941K Followers 275 FollowingWe’re a team of scientists, engineers, ethicists and more, committed to solving intelligence, to advance science and benefit humanity.
40K Followers 1K FollowingSecular Bayesian.
Associate Professor in Machine Learning @Cambridge_CL. Talent aficionado at https://t.co/RbJkoLguey
Alum of @Twitter, Magic Pony and @Balderton
45K Followers 2K FollowingResearch Director, @VectorInst. Canada CIFAR AI Chair. Associate Professor of Stats/CS @UofT. I study machine learning and AI, emphasis on theory.
111K Followers 35 FollowingNew Orleans, Dec 10-16, 23. https://t.co/ga8aOw615g Tweets to this account are not monitored. Please send feedback to [email protected].
689 Followers 335 FollowingPhD student @OxfordStats / @OXCSML supervised by @yeewhye and @wellingmax.
Probabilistic ML, geometric ML and their interestion.
Interned @DeepMind @Qualcomm
60K Followers 2K Followinga combination of a mediocre scientist, a mediocre manager, a mediocre advisor & a mediocre PC at @nyuniversity (@CILVRatNYU) & @genentech (@PrescientDesign).
36K Followers 773 FollowingFR/US/GB AI/ML Person, Director of Research at @GoogleDeepMind, Honorary Professor at @UCL_DARK, @ELLISforEurope Fellow. All posts are personal.
22K Followers 1K FollowingKnowledge manifests itself in radiant dreams that shimmer like the wild sun
Views are my own
pfau at sigmoid dot social on 🦣
https://t.co/xqtVHHVI17 on 🦋
807 Followers 292 FollowingAsst. prof at @Yale CS. I work on robots that learn from people 🤖
Formerly postdoc @CMU_Robotics, PhD @GeorgiaTech, BSc @Portland_State. (she/her)
5K Followers 600 Following•Researcher in theoretical CS+statistics applied to healthcare @broadinstitute @Schmidt_Center
•PhD in probabilistic ML @Cambridge_Uni
•Founder @AiDisability
806K Followers 962 FollowingMenswear writer. Editor at @putthison. Creator of @RLGoesHard. Bylines at The New York Times, The Washington Post, The Financial Times, Esquire, and Mr. Porter
4K Followers 847 Followinghow shall we live together?
co-founder @collect_intel ⋅ research @ uk ai safety institute
was @googledeepmind • co-created @kernel_magazine ⋅ TK
453K Followers 11K FollowingTrump Posts From Truth Social & Related So They Can Be Shared With Millions On X. Make Sure To Tap (🔔Notify+) So Tweets Cannot Be Suppressed. No Affiliation.
2K Followers 684 FollowingGoogle PhD Fellow at @uncnlp. Interested in interpretable ML, natural language processing, AI Safety, and Effective Altruism.
644 Followers 633 Followinga neutral networker. a language lover. Senior ML Scientist at @awscloud. Previous - @MSFTResearch, @Cornell PhD. https://t.co/15TmafCLB1
425 Followers 130 FollowingResearch fellow @FlatironCCM and @mcmc_stan developer. Interested in computational and applied statistics. Opinions are my own.
445 Followers 6 FollowingJoin us in discovering how generative AI can revolutionize protein research, RNA analysis, molecular design, drug discovery, and more.
3K Followers 452 FollowingPostdoc at @MSFTResearch | Assistant Professor at @UChicagoCS (2024) | PhD at @Stanford | Language models, AI-assisted writing, Human-AI interaction ✍️
1K Followers 965 FollowingComputer Science Ph.D. Student @ Stanford | Responsible AI Research Lead @ Stanford AI Index | Responsible AI | AI Governance | Views are my own
2K Followers 1K FollowingOn the academic job market! CS PhD student @Harvard working on ML and decision making for planetary health. Bridging research and practice through @MD4SG.
2K Followers 655 FollowingAI researcher trying to make sense of all things cyberspace 🤖 Uni of Ox PhD (loading…) @oiioxford & @OxfordAI. Prev @turinginst & @Cambridge_Uni. Visitor @ NYU
162K Followers 745 FollowingThe official Twitter feed for Yale School of Medicine, a world-renowned center for biomedical research, education, and health care founded in 1810.