Andrew Ilyas @andrew_ilyas
Machine Learning PhD student at MIT, advised by Aleksander Madry and Costis Daskalakis. andrewilyas.com Joined February 2013-
Tweets252
-
Followers2K
-
Following167
-
Likes531
Our second (and final) blog post on model components is now out: we show that component attributions enable model editing! See the blog post for more: gradientscience.org/modelcomponent… Paper: arxiv.org/abs/2404.11534 (this time with no typos :) Code: github.com/MadryLab/model…
Our second (and final) blog post on model components is now out: we show that component attributions enable model editing! See the blog post for more: gradientscience.org/modelcomponent… Paper: arxiv.org/abs/2404.11534 (this time with no typos :) Code: github.com/MadryLab/model…
New work with @andrew_ilyas and @aleks_madry on tracing predictions back to individual components (conv filters, attn heads) in the model! Paper: arxiv.org/abs/2404.11534 Thread: 👇
New work with @andrew_ilyas and @aleks_madry on tracing predictions back to individual components (conv filters, attn heads) in the model! Paper: arxiv.org/abs/2404.11534 Thread: 👇
New work led by @harshays_ on studying how model components affect predictions! Paper link: arxiv.org/abs/2404.11534 Blog: gradientscience.org/modelcomponents
New work led by @harshays_ on studying how model components affect predictions! Paper link: arxiv.org/abs/2404.11534 Blog: gradientscience.org/modelcomponents
Congratulations @ShibaniSan, this is extremely well deserved!
Congratulations @ShibaniSan, this is extremely well deserved!
We are pleased to announce the 2021 AAAI/ACM SIGAI Dissertation Award Winner. Congratulations to Shibani Santurkar, Massachusetts Institute of Technology for her work entitled Machine Learning Beyond Accuracy: A Features Perspective On Model Generalization. And congratulations…
We tend to choose LM training data via intuitive notions of text quality... but LMs are often *un*intuitive. Is there a better way? w/@logan_engstrom, @axel_s_feldmann: we select better data by modeling how models learn from data. Our method, DsDm, can greatly improve…
How do we attribute an image generated by a diffusion model back to the training data? w/ @kris_georgiev1 @josh_vendrow @hadisalmanX @smsampark we show that it’s useful to look at each step of the diffusion process:
I gave a keynote this week at the fantastic ATTRIB Workshop #NeurIPS2023 "What does scale give us: Why we are building a ladder 🪜 to the moon 🌕" Some of you asked for my slides, sharing below: docs.google.com/presentation/d… Thanks to the organizers for a fantastic workshop! 🔥
Next - @sarahookr on understanding the effects of scale and data on ML model performance! (ATTRIB workshop, Rm 271-273)
What types of attributions do modern LLM applications require? Check out our contributed talk [Friday, 10:30am] by @TeddiWorledge at the ATTRIB23 workshop [Rm 271-273] on "Unifying Corroborative and Contributive Attributions in Large Language Models" arxiv.org/abs/2311.12233
Any burning ML questions? The ATTRIB workshop is hosting a panel on "The Future of Attribution in ML" tomorrow at 11AM and is soliciting questions! Submit them by TODAY 11:59PM to hear answers at the panel tomorrow! forms.gle/Yd5N3Ti6kKfqij… More info: attrib-workshop.cc
Gautam Kamath @thegautamkamath
44K Followers 505 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Clément Canonne @ccanonne_
31K Followers 928 Following Senior Lecturer @Sydney_Uni. Postdocs @IBMResearch, @Stanford; PhD @Columbia. Converts ☕ into puns: sometimes theorems. He/him. @[email protected]Dan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)Aleksander Madry @aleks_madry
31K Followers 165 Following Head of Preparedness at OpenAI and MIT faculty (on leave). Working on making AI more reliable and safe, as well as on AI having a positive impact on society.Behnam Neyshabur @bneyshabur
18K Followers 689 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingAnanya Kumar @ananyaku
4K Followers 469 Following Researcher at @openai Previously PhD at Stanford University (@StanfordAILab) advised by Percy Liang and Tengyu MaZachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷Jeremy Cohen @deepcohen
4K Followers 868 Following PhD student in machine learning at Carnegie Mellon. The goal of my research is to turn deep learning into a real engineering discipline.Alex Dimakis @AlexGDimakis
13K Followers 2K Following UT Austin Professor. Researcher in Machine Learning and Information Theory. National AI Institute on the Foundations of Machine Learning (IFML) Co-director.Sara Hooker @sarahookr
39K Followers 7K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.Tom Goldstein @tomgoldsteincs
23K Followers 2K Following Professor at UMD. AI security & privacy, algorithmic bias, foundations of ML. Follow me for commentary on state-of-the-art AI.Miles Brundage @Miles_Brundage
43K Followers 10K Following Policy research at @openai. I mostly tweet about AI, animals, and sci-fi. He/him. Views my own.Jeff Dean (@🏡) @JeffDean
296K Followers 6K Following Chief Scientist, Google DeepMind and Google Research. Co-designer/implementor of things like @TensorFlow, MapReduce, Bigtable, Spanner, Gemini .. (he/him)Amin Karbasi @aminkarbasi
8K Followers 2K Following Associate Professor at Yale University, staff research scientist at Google.Jason Lee @jasondeanlee
10K Followers 3K Following Associate Professor at Princeton and Research Scientist at Google DeepMind. ML/AI Researcher working on foundations of LLMs and deep learningSam Power @sp_monte_carlo
17K Followers 7K Following Lecturer in Maths & Stats at Bristol. Interested in probabilistic + numerical computation, statistical modelling + inference. (he / him)Ethan Caballero is bu.. @ethanCaballero
8K Followers 2K Following ML PhD student @Mila_Quebec ; previously @GoogleDeepMindTengyu Ma @tengyuma
25K Followers 512 Following Assistant professor at Stanford; Co-founder of Voyage AI (https://t.co/wpIITHLgF0) ; Working on ML, DL, RL, LLMs, and their theory.Greg Yang @TheGregYang
53K Followers 661 Following Cofounder https://t.co/SpHbO7FZNV. Morgan Prize Honorable Mention 2018. Developing the theory of #TensorPrograms and the practice of scaling #neuralnetworks.Huy Tran @huytransformer
92 Followers 3K FollowingB @bbbb_bb_b
0 Followers 3K FollowingYating Wu @YatingWu96
142 Followers 188 Following ECE Ph.D student @ UT Austin, advised by @jessyjli and @AlexGDimakis | テキサス大学に在籍する博士生Return of JC @A76499780
198 Followers 934 FollowingRyon @taxi1729_ryon
496 Followers 2K Following 京大情報B4 AtCoder 最高水色1506 計算機科学コース SecHack365研究駆動(2023)Leo Zhu @Leo_Bayes
35 Followers 96 FollowingNic Fishman @njwfish
885 Followers 803 Following ever wasting time on the next new thing. prev: cs/soc @stanford, ml @oxfordstats. now: stats @harvard. always @zahra_thab. he/they.Zhaoyang Wang @wangwan83764204
300 Followers 4K Following CS PhD student at UoB in the United Kingdom. Research interests: Automated Machine Learning, Online Learning, and Reinforcement Learning 🏳️🌈Vaibhav Dixit @Vaibhavdixit02
218 Followers 1K Following Optimization, Bayesian Inference, Scientific Machine Learning and #julialangIbrahim Ahmad @Ibrahim63433664
85 Followers 3K FollowingManuel Gomez-Rodrigue.. @autreche
1K Followers 136 Following Human-centric machine learning at the Max Planck Institute for Software Systems.Ujir Yed @UjirYed
20 Followers 45 FollowingWilliam , Engineer, C.. @whe3998
3K Followers 5K Following AI, ML,EV,ELON and MORE. NOT investment Advice.Arif Ahmad @arif_ahmad_py
248 Followers 7K Following All things AI, Computer Science and Circuits! Prev. @GoogleAIAbhigyan Bhusal @AbhigyanBhusal
44 Followers 160 FollowingJack Jingyu Zhang @jackjingyuzhang
266 Followers 631 Following PhD student @jhuclsp | previously undergrad @JohnsHopkins @JHUCompSci.Alexandra DeLucia @Alexir563
548 Followers 740 Following Sony intern Fall 2023. Computer Science PhD Student at @jhuclsp in the @mdredze group. @rollinscollege alum.Jess Sorrell @JessSorrell
207 Followers 399 Following CS postdoc at Penn. Interested in theory of ML and responsible computing. All cat pictures are my own and do not represent the cats of my employer.Pingbang Hu 🇹🇼 @PingbangHu
61 Followers 112 Following Ph.D. student @UofIllinois. Incoming Research Intern @jouhouken. Previously @Umich & @SJTU1896. I give theoretical insight to AI with math on graphs.Sourjya Sarkar @SarkarSourjya
39 Followers 964 Following Loves traveling, music, photography, friends, research and fun !! A bit of a machine learning enthusiast!Pratinav Seth @ptnv_s
395 Followers 3K Following AAAI-UC Scholar 23 | Intern @Mila_Quebec , Prev- @Bosch_Research ,KLIV IITKGP | Looking for Full Time AI/DS/ML/RS Roles |Akshit Khatgarh @AKhatgarh
31 Followers 1K Followingkumar @kumar__nn
0 Followers 1K FollowingBhavaniKola @BkolaMath
65 Followers 227 Following Bhavani Kola. Boy Mom. Passionate Educator #MIEMasterTrainer #MIEExpert20 #AdobeCreativeEducator #YoutuberYuntao Du @scott_ytdu
8 Followers 53 Following CS PhD student at @PurdueCS. Data privacy and AI security.Katherine Xu @KateYXu
164 Followers 918 Following @Penn PhD student • @MIT CS '22 🤺 • computer vision + generative AIRebecca Lin @rebeccayelin
222 Followers 182 Following artist and researcher phding @mit_csail, learning about shapes and structuresJitendra Sharma @jkumarsharma998
808 Followers 6K Following Curious about Research in AI. NLP and Computer Vision Interest me. Curious about truth and existence. Views are personal.Ekaterina Lobacheva @KateLobacheva
300 Followers 204 Following Independent Researcher, PhD in CS, Collaboration with https://t.co/tB7QL7Sw3a and https://t.co/JN95AWiNhB Like to explain unexpected behavior of neural nets 🤯Bartosz Cywinski @bartoszcyw
17 Followers 444 FollowingZoky @SkivenK
32 Followers 34 FollowingGautam Kamath @thegautamkamath
44K Followers 505 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Clément Canonne @ccanonne_
31K Followers 928 Following Senior Lecturer @Sydney_Uni. Postdocs @IBMResearch, @Stanford; PhD @Columbia. Converts ☕ into puns: sometimes theorems. He/him. @[email protected]Dan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)Percy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistAleksander Madry @aleks_madry
31K Followers 165 Following Head of Preparedness at OpenAI and MIT faculty (on leave). Working on making AI more reliable and safe, as well as on AI having a positive impact on society.Peyman Milanfar @docmilanfar
67K Followers 262 Following Distinguished Scientist at Google Research. Computational Imaging, Machine Learning, and Vision. Tweets = personal opinions. May change or disappear over time.Zachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷Boaz Barak @boazbaraktcs
17K Followers 419 Following Computer Scientist. See also https://t.co/EXWR5k634w, https://t.co/SEVX6it6z3 ( @[email protected] , boaz.barak in threads ). Opinions my own.Jeremy Cohen @deepcohen
4K Followers 868 Following PhD student in machine learning at Carnegie Mellon. The goal of my research is to turn deep learning into a real engineering discipline.Alex Dimakis @AlexGDimakis
13K Followers 2K Following UT Austin Professor. Researcher in Machine Learning and Information Theory. National AI Institute on the Foundations of Machine Learning (IFML) Co-director.Sara Hooker @sarahookr
39K Followers 7K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.Tom Goldstein @tomgoldsteincs
23K Followers 2K Following Professor at UMD. AI security & privacy, algorithmic bias, foundations of ML. Follow me for commentary on state-of-the-art AI.Jeff Dean (@🏡) @JeffDean
296K Followers 6K Following Chief Scientist, Google DeepMind and Google Research. Co-designer/implementor of things like @TensorFlow, MapReduce, Bigtable, Spanner, Gemini .. (he/him)Jason Lee @jasondeanlee
10K Followers 3K Following Associate Professor at Princeton and Research Scientist at Google DeepMind. ML/AI Researcher working on foundations of LLMs and deep learningMaggie Makar @Maggiemakar
4K Followers 628 Following Assistant prof @umichcse. Previously @MIT_CSAIL. Machine learning. Causal models. Healthcare. Swimming. @[email protected] Opinions are my own.Tengyu Ma @tengyuma
25K Followers 512 Following Assistant professor at Stanford; Co-founder of Voyage AI (https://t.co/wpIITHLgF0) ; Working on ML, DL, RL, LLMs, and their theory.Pravesh K. Kothari @praveshkkothari
3K Followers 294 Following Assistant Professor @PrincetonCS, @the_ias Research: Theoretical Computer Science, Optimization, Algorithmic Statistics.Manuel Gomez-Rodrigue.. @autreche
1K Followers 136 Following Human-centric machine learning at the Max Planck Institute for Software Systems.Rebecca Lin @rebeccayelin
222 Followers 182 Following artist and researcher phding @mit_csail, learning about shapes and structuresJohan Ugander @jugander
6K Followers 1K Following Associate Professor, Stanford MS&E. Social networks, social and behavioral data. Occasionally disappear into the mountains. Other place: juganderJohn Cherian @jjcherian
2K Followers 352 Following Statistics PhD student @Stanford and @HertzFoundation fellow | Quantifying uncertainty @WashingtonPost | Formerly at @DEShawResearchAxel Feldmann @axel_s_feldmann
77 Followers 204 Following PhD student @mit_csail studying computer architectureNikhil Garg @NikhGarg
2K Followers 667 Following I study algorithms/learning/data applied to democracy/markets/society. Asst. professor @cornell_tech @Cornell_ORIE. "nkgar6" on threadsTeddi Worledge @TeddiWorledge
37 Followers 80 Following (she/her) Computer Science PhD Student @Stanford. Formerly @Berkeley.Katherine Lee @katherine1ee
6K Followers 932 Following understanding ourselves and our models. senior research scientist @GoogleBrain, @genlawcenter and @CornellCIS, formerly @Princeton @[email protected]Benjamin Laufer @BenDLaufer
459 Followers 510 Following PhD student @cornellCIS based in NYC @Cornell_Tech. Formerly ORFE @Princeton.Princewill Okoroafor @pc_okoroafor
74 Followers 128 Following CS Theory PhD Student @Cornell | @HarveyMudd '20 | @ALAcademy '16Hari Balakrishnan @HariBalakrish20
4K Followers 181 Following Technologist. Founder & CTO of Cambridge Mobile Telematics. Professor at MIT CSAIL. Mobility, Networks, Sensing, Perception, Systems. Cricket.Susan Zhang @suchenzang
20K Followers 503 Following @ Google Deepmind. Past: @MetaAI, @OpenAI, @unitygames, @losalamosnatlab, @Princeton etc. Always hungry for compute.Vasilis Syrgkanis @syrgkanis
920 Followers 65 Following Asst. Professor, Stanford MS&E. Ex-Principal Researcher at Microsoft Research, working on machine learning, econometrics, causal inference, and game theory.Brown NLP @Brown_NLP
3K Followers 144 Following Language Understanding and Representation Lab at Brown University. PI: Ellie Pavlick.Monica Agrawal @MonicaNAgrawal
3K Followers 305 Following Incoming asst prof at @DukeU (July 2024). Co-founder at @LayerHealth. ML and NLP for healthcare. PhD @mit_csail, BS/MS @stanford. She/her.Yoav Wald @wald_yoav
233 Followers 141 Following Faculty fellow @NYUDataScience. ML safety, causal inference, healthcare applications.Joel Flynn @joel_flynn
862 Followers 405 Following Macro and Mechanisms. Joining Yale as a Post-Doc (23-24) and then as an Assistant Professor (24-).Rohan Alur @rohanalur
151 Followers 230 Following PhD student @MITEECS. Interested in machine learning, causal inference and human/AI interaction.Jenny Davis @Jenny_L_Davis
5K Followers 5K Following Sociologist at Vanderbilt | Critical Tech+Social Psych| Book: https://t.co/oc7NZzqqz6|Christopher Jung @crispy_jung
303 Followers 355 Following Postdoc at @stanford | Previously @penn phd, @cornell | ChristianHarshay Shah @harshays_
406 Followers 450 Following ML PhD student at MIT, advised by @aleks_madry Previously: @googleai @msftresearch @illinoiscsKeller Jordan @kellerjordan0
1K Followers 197 Following Independent research Prev MLE @ Hive AI, math @ UCSDSeong Joon Oh @coallaoh
1K Followers 869 Following Leading the STAI group at the University of Tübingen https://t.co/qrSPDDcdOy Advising @ParameterLab.Sam Park @smsampark
314 Followers 347 Following Machine Learning PhD student at MIT, advised by Aleksander Mądry.Isabella Struckman @imstruckman
41 Followers 12 Following Senior undergrad @MIT studying AI & Decision Making and concentrating in EthicsTatsunori Hashimoto @tatsu_hashimoto
6K Followers 202 Following Assistant Prof at Stanford CS, member of @stanfordnlp and statsml groups; Formerly at Microsoft / postdoc at Stanford CS / Stats.Luis Videgaray Caso @LVidegaray
1.2M Followers 644 Following Senior Lecturer, MIT Sloan. Director, MIT AI Policy for the World Project. Co-lead, MIT AI Policy Forum.Awais @deepawais
1K Followers 5K Following Researcher @MBZUAI Prev: @SonyAI_global, Jp 🇯🇵; Noah's Ark Lab, @Huawei, HK 🇭🇰; MLVC Lab, KHU, SK 🇰🇷Shengwu Li @ShengwuLi
35K Followers 760 Following Economist. @Harvard. Mechanism design, behavioral theory.Hannah Li @hannahq_li
781 Followers 414 Following Assistant prof @Columbia_Biz | Previously @Stanford @MIT @PomonaCollege | Thinking about online platforms, A/B testing, and educationPaul Poast @ProfPaulPoast
95K Followers 473 Following Was `Tweeting to teach', now `Posting for pedagogy'. International Relations and Foreign Policy. @UChicago Prof. @ChicagoCouncil Fellow. @WPReview Columnist.Elan Rosenfeld @ElanRosenfeld
1K Followers 184 Following Final year ML PhD Student at CMU working on principled approaches to robustness, security, generalization, and representation learning.Stephen McAleer @McaleerStephen
3K Followers 807 Following Postdoc at CMU researching LLM agents and AI alignmentRuiqi Zhong @ZhongRuiqi
2K Followers 698 Following 5th Year Ph.D. @BerkeleyNLP, Columbia'19. part time working for @AnthropicAI . Supervising machines to do what I can't do.Kristian Georgiev @kris_georgiev1
347 Followers 599 Following PhD at @MIT, advised by @aleks_madry | previously: @Apple & @quantcotechYann Dubois @yanndubs
4K Followers 1K Following PhD student @stanfordAILab | Prev: AI resident @metaai, @vectorinst, @CambridgeMLGChara Podimata @charapod
744 Followers 367 Following Assistant Professor of OR/Stat at MIT. Past: postdoc UC Berkley, PhD Harvard. Interested in incentive-aware ML, bandits, mechanism design and public policy.MIT CSAIL @MIT_CSAIL
298K Followers 22K Following MIT's Computer Science & Artificial Intelligence Laboratory (CSAIL). Media Inquiries: [email protected]Ziad Obermeyer @oziadias
7K Followers 689 Following Physician + prof @UCBerkeley. Co-founder, @NightingaleOS, @Dandelion_AI4H. Machine learning, health, emergencies, espresso.Elizabeth Yang @eytyang
768 Followers 677 Following technical staff @openai, previously theory @berkeleyeecs | fan of graphs, crosswords, turtles, bad puns, running, and Survivor, among other thingsTheo Diamandis @theo_diamandis
3K Followers 826 Following optimization phd @mit. head intern @baincapcryptoProud Advisor Moment: Giannis passes his Phd Proposal with flying colors. You can see this great thread about (some of) his work. The new in-painting attack for checking if an image is in the training set of a diffusion model is one of my favorites.
This week I successfully passed my Ph.D. proposal 🎉 The title of the talk: "Generative Models from Lossy Measurements". Here is a little 🧵 about it
HELM Lite v1.2.0 is out! Datasets: NarrativeQA, NaturalQA, OpenbookQA, MMLU, MATH, GSM8K, LegalBench, MedQA, WMT14 Results (we still need to add Claude 3, which requires more prompt finagling): crfm.stanford.edu/helm/lite/v1.2…
Our second (and final) blog post on model components is now out: we show that component attributions enable model editing! See the blog post for more: gradientscience.org/modelcomponent… Paper: arxiv.org/abs/2404.11534 (this time with no typos :) Code: github.com/MadryLab/model…
How do model components (conv filters, attn heads) collectively transform examples into predictions? Is it possible to somehow dissect how *every* model component contributes to a prediction? w/ @harshays_ @andrewilyas, we introduce a framework for tackling this question!…
New work with @andrew_ilyas and @aleks_madry on tracing predictions back to individual components (conv filters, attn heads) in the model! Paper: arxiv.org/abs/2404.11534 Thread: 👇
How do model components (conv filters, attn heads) collectively transform examples into predictions? Is it possible to somehow dissect how *every* model component contributes to a prediction? w/ @harshays_ @andrewilyas, we introduce a framework for tackling this question!…
How do model components (conv filters, attn heads) collectively transform examples into predictions? Is it possible to somehow dissect how *every* model component contributes to a prediction? w/ @harshays_ @andrewilyas, we introduce a framework for tackling this question!…
We can train a diffusion using only these noisy images and produce natural looking images. New results from our paper "Consistent Diffusion Meets Tweedie"
Consistent Diffusion Meets Tweedie. Our latest paper introduces an exact framework to train/finetune diffusion models like Stable Diffusion XL solely with noisy data. A year's worth of work breakthrough in reducing memorization and its implications on copyright 🧵
Can you train a generative model using only noisy data? If you can, this would alleviate the issue of training data memorization plaguing certain genAI models. In exciting work with @giannis_daras and @AlexGDimakis we show how to do this for diffusion-based generative models.
Consistent Diffusion Meets Tweedie. Our latest paper introduces an exact framework to train/finetune diffusion models like Stable Diffusion XL solely with noisy data. A year's worth of work breakthrough in reducing memorization and its implications on copyright 🧵
TacticAI is an AI system that can advise football coaches on tactics & plays deepmind.google/discover/blog/… This was a really fun project to work on in collaboration with my much loved Liverpool FC @LFC - fingers crossed we win the league this year to give Klopp a fitting send off!
When we looked for a final strategic partner to fulfill our dreams for series C, only one name came to mind: @nvidia. We are so thrilled to welcome our newest investors, collaboration partners, and long-time research buddies to the @AbridgeHQ family: abridge.com/press-release/…
Congratulations @ShibaniSan, this is extremely well deserved!
We are pleased to announce the 2021 AAAI/ACM SIGAI Dissertation Award Winner. Congratulations to Shibani Santurkar, Massachusetts Institute of Technology for her work entitled Machine Learning Beyond Accuracy: A Features Perspective On Model Generalization. And congratulations…
We are pleased to announce the 2021 AAAI/ACM SIGAI Dissertation Award Winner. Congratulations to Shibani Santurkar, Massachusetts Institute of Technology for her work entitled Machine Learning Beyond Accuracy: A Features Perspective On Model Generalization. And congratulations…
Excited to share our new applied causal machine learning book causalml-book.org is available online. Any feedback/corrections greatly appreciated!
It's such an honor to be considered in the company of the many brilliant people that have been promoted to the rank of associate professor. #humblebrag
What shall it be Twitter: platitude, hot take, or humble brag? #itakerequests
This subset is hard to find directly–we typically optimize loss with respect to parameters, not training data! Instead, DsDm approximates the solution using datamodels (arxiv.org/abs/2202.00622), a framework for modeling how ML models predict from data.
We tend to choose LM training data via intuitive notions of text quality... but LMs are often *un*intuitive. Is there a better way? w/@logan_engstrom, @axel_s_feldmann: we select better data by modeling how models learn from data. Our method, DsDm, can greatly improve…
I agree with Percy's call to arms for open models, I think it will be very important for our world. As investments for academic compute are growing, what is lagging the most is high quality datasets. The academic and open source community must work more on dataset curation and…
My TEDAI talk from Oct 2023 is now live: go.ted.com/percyliang It was a hard talk to give: 1. I memorized it - felt more like giving a piano recital than an academic talk. 2. I wanted it to be timeless despite AI changing fast…still ok after 3 months. Here’s what I said:
How do we attribute an image generated by a diffusion model back to the training data? w/ @kris_georgiev1 @josh_vendrow @hadisalmanX @smsampark we show that it’s useful to look at each step of the diffusion process:
I gave a keynote this week at the fantastic ATTRIB Workshop #NeurIPS2023 "What does scale give us: Why we are building a ladder 🪜 to the moon 🌕" Some of you asked for my slides, sharing below: docs.google.com/presentation/d… Thanks to the organizers for a fantastic workshop! 🔥
Join us at the panel where we discuss the big picture for attribution and explaining model behavior.
Up next at ATTRIB (in 5 minutes): A panel on the future of Attribution in ML, with @PangWeiKoh, @sarahookr, and @katherine1ee ! Room 271-273
🔥🎉 @maxdoesresearch presents “when less is more: investigating data pruning for pretraining LLMs at scale” Attrib Workshop 2023