Découvrez Les IA Experts
Nando de Freitas | Chercheur chez Deepind | |
Nige Willson | Conférencier | |
Ria Pratyusha Kalluri | Chercheur, MIT | |
Ifeoma Ozoma | Directrice, Earthseed | |
Will Knight | Journaliste, Wired |
Nando de Freitas | Chercheur chez Deepind | |
Nige Willson | Conférencier | |
Ria Pratyusha Kalluri | Chercheur, MIT | |
Ifeoma Ozoma | Directrice, Earthseed | |
Will Knight | Journaliste, Wired |
Profil AI Expert
Non Disponible
Les derniers messages de l'Expert:
2024-03-01 00:00:00 CAFIAC FIX
2024-03-11 00:00:00 CAFIAC FIX
2023-05-20 05:43:25 RT @GoogleAI: Today, we discuss the current state of differentially private ML (DP-ML) research with an overview of common techniques for o…
2023-05-19 19:00:00 CAFIAC FIX
2023-05-21 19:00:00 CAFIAC FIX
2023-04-21 00:00:01 CAFIAC FIX
2023-04-15 03:04:25 @TaliaRinger A very nice profile of your many different threads of work!
2023-04-15 02:32:09 RT @SashaMTL: Our @TechaideMTL conference raised 100,000 dollars for charity today. I couldn't have asked for a better birthday present. Am…
2023-04-15 01:39:08 @kristintynski @emilymbender I tried. Note my email from 20 months ago.
2023-04-14 07:02:56 @vote_no_body Perhaps you can explain why congratulating someone on well deserved recognition of their work and the person appreciating that makes you react this way?
2023-04-14 06:28:09 @timnitGebru @vote_no_body @emilymbender https://t.co/HvqMl9l7ye
2023-04-14 05:51:43 @emilymbender Also, I reached out to you to discuss ~20 months ago (following your directions on how to email you from your home page) and never heard back. 'Please include the word "penguins" in the subject line so that I know that you've read this page.' https://t.co/CnqAUN9EFk
2023-04-14 02:47:54 @emilymbender Given that Meg liked my reply, and Twitter is a public space, I'll presume my earnest congratulations are fine.
2023-04-13 18:12:21 @mmitchell_ai Congrats!
2023-04-09 19:59:16 @DynamicWebPaige https://t.co/qC7TewtF0I
2023-04-09 19:58:31 @DynamicWebPaige Fun fact: a huge amount of the west coast's internet traffic goes through this building (or did a decade or two ago), despite its rather modest "nice historical building" appearance and location a couple doors down from a popular ramen spot.
2023-04-08 04:04:26 @kevin_zakka Rack your brain to rach your robo brain to rock the world. Caught between a rach and a hard piece.
2023-04-05 02:09:24 Paper: TPUv4 system has an optically reconfigurable network to assemble groups of 4x4x4 chips like legos (4x4x12? 16x16x16?). SparseCores help w/ embeddings. TPUv4 outperforms TPUv3 by 2.1x &
2023-04-05 01:53:59 RT @ZetaVector: In the recent discussion about the impact of AI R&
2023-04-02 21:11:03 RT @0xDesigner: Probably the best thing you'll see today. In 2017, a group of developers hilariously competed for who could create worst v…
2023-03-30 23:32:21 RT @kevin_zakka: Introducing , a new benchmark for high-dimensional robot control! Solving it requires mastering the piano wi…
2023-03-28 00:07:52 @DH_PlantVillage Liking for the visibility people should have into this data, not for the increasing trend.
2023-03-26 01:49:40 @NLPurr I found your post interesting. Thanks for the analysis.
2023-03-25 15:55:08 @Rubenia_Borge Congrats! One never forgets one's first research talk!
2023-03-24 23:02:46 @strubell @vijaytarian @_jasonwei Yes, I totally agree that the fact that the proxy task was Transformer(base) is quite subtle in So et al. (just that one sentence). NAS is generally a once-per-search-space/problem-type (sort of like human researchers coming up with a new model architecture, just automated).
2023-03-24 07:41:43 @strubell @vijaytarian @_jasonwei As I said above, use of Transformer-Base as proxy task *is* in So et al: "Specifically, to train a Transformer to peak performance on WMT’14 En-De requires ∼300K training steps, or 10 hours, in the base size when using a single Google TPU V.2 chip, as we do in our search"
2023-03-22 23:16:03 RT @PrecursorCell: Dear Reviewer 1 of our manuscript, You are my hero. Sincerely, A very drained PhD student https://t.co/AAyny1QVJL
2023-03-22 10:38:25 Google is proud to sponsor the Turing Award, to help recognize great computer science work.
2023-03-22 10:37:10 The original Ethernet paper is well worth a read if you haven't seen it: https://t.co/7SgENryibW
2023-03-22 10:34:30 Congrats to @RobertMMetcalfe as the winner of the 2022 Turing Award! His pioneering work on Ethernet literally helps connect us all. https://t.co/roOFsVESbR
2023-03-22 01:02:55 RT @TheGradient: Applications are open for the @Google Carbon Removal Research Awards, a program looking to fund selected academic research…
2023-03-21 22:33:00 @TaliaRinger I profess that it is and you can (and you should: sounds like a great class to take from you!)
2023-03-21 14:41:20 Bard is now available in the US and UK, w/more countries to come. It’s great to see early @GoogleAI work reflected in it—advances in sequence learning, large neural nets, Transformers, responsible AI techniques, dialog systems &
2023-03-20 09:14:30 @penberg @tim_zaman I gave a talk at an MIT workshop on the Climate Impacts of Computing and Communication a year or so ago that touches on some higher level issues in this space like provisioning renewable energy, emissions of ML, ... (first 30 minutes of this video): https://t.co/edHaYXWWHw
2023-03-19 22:05:50 @lifesafeast That's peachy! Congrats!
2023-03-19 22:01:18 @gazpachodragon @kbhoyle_author Glad I mustard the courage to post it, then!
2023-03-19 21:30:24 @kbhoyle_author Don't poupon the Dijon raincoat! It's great!
2023-03-19 21:27:59 @penberg @labarroso @uhoelzle Also, recent editions of my other colleagues Dave Patterson and John Hennessy's classic textbook Computer Architecture: A Quantitative Approach have a chapter or two of relevant content.
2023-03-19 20:51:58 @__apf__ Time to dust off something else to worry about!
2023-03-19 20:50:09 @thingskatedid Your uncle sounds like a really wonderful, positive-outlook, progressive, curious about the world and its wonders and changes kind of person, and thank you for sharing. People like this can make us all happy and better.
2023-03-19 20:41:11 @penberg I'll plug the book by my colleagues @labarroso and @uhoelzle. The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines (Synthesis Lectures on Computer Architecture) https://t.co/35eLcB93En
2023-03-19 09:04:47 RT @Ishwariya13: My dad left us 7 years ago today. He quit his work &
2023-03-19 08:59:26 @Ishwariya13 What a nice tribute to your dad! He would be so proud.
2023-03-18 21:34:12 @ayyar @the_transit_guy I really like that movie. I don't know how accurately it portrays Mozart, but it gives some vivid movie memories to particular Mozart pieces.
2023-03-18 21:15:27 @MatejVela Glad you liked it!
2023-03-16 00:25:55 A bit of a peek into the PaLM LLM APIs we announced yesterday (currently in private preview with a few customers/users to get feedback: we'll broaden access in the coming weeks) https://t.co/KYpIbWy3ij
2023-03-15 04:15:14 @RanthonyEdmonds Congrats! Happiness is finding a partner to share a slice of life with, especially on π Day!
2023-03-14 16:45:08 You can find more info about Med-PaLM2 and other work we discussed today at #TheCheckup in this blog post. https://t.co/IkjOsRWAzd
2023-03-14 16:36:26 In December, we discussed Med-PaLM, at that time a SOTA medical LLM that achieved a 67.6% score on the USMLE MedQA evaluation (passing is 60%). Today, we're describing Med-PaLM2, which improves on this by +18% with a score of 85.4% ("expert performance")! Kudos to all involved! https://t.co/nlNxkm50Up
2023-03-14 16:22:59 RT @Google: Tune in now to the 3rd annual #TheCheckUp with @GoogleHealth to learn about our latest advancements in health and how we’re hel…
2023-03-14 16:22:20 RT @GoogleOSS: Meet Open Health Stack! It’s a suite of open source tools that uses FHIR to make it easier to build interoperable st…
2023-03-14 13:56:10 Excited to share a number of features that build on many years of @GoogleAI research and generative AI advances: An API to the latest generation of our PaLM language models MakerSuite for developers, Writing/rewriting help in Docs and Gmail (Google Workspace), and more! https://t.co/5Fk3PIUsus
2023-03-14 05:56:52 @BenTheEgg As an extreme version: taking work that was actually run on EfficientAcceleratorsA in SolarPowerDataCenterB &
2023-03-14 05:47:33 @BenTheEgg Because carbon emissions is highly dependent on which data center you run in, which electricity grid it is powered by, etc. If you want to talk about the amount of computation needed to train one model architecture vs. another, that seems reasonable to compare abstractly. ...
2023-03-14 03:43:09 @pcastr Buenos Aires is such an amazing city! Lovely photos. I was very fortunate to spend a few days there after @Khipu_AI 2019.
2023-03-14 03:33:42 @vijaytarian @_jasonwei @strubell The use of smaller proxy tasks was covered in Section 3.3 of So et al., so no Google inside knowledge needed (although it's pretty subtly described in So et al.) There's some discussion in Appendix C of https://t.co/W0fNJGCV6c
2023-03-14 03:20:01 @vijaytarian @_jasonwei @strubell Presenting the one-time cost of a NAS that has already been conducted in an energy efficient environment (and the best architecture from it open sourced) as if it was an every problem cost is pretty confusing to the reader. "Counting building the factory for every car produced."
2023-03-14 03:05:37 RT @forai_ml: We’re still buzzing from the energy of @Khipu_AI and wanted to share some of our favourite moments from a special week in Mon…
2023-03-13 05:09:27 I'm sorry I wasn't able to make it to @Khipu_AI this year. I had such an amazing time &
2023-03-13 04:39:39 @geofflangdale It's actually more ironic. The results of one-time NAS (which took 88X less CO2 than the external paper estimated) are open-sourced and make training language models 1.3X faster &
2023-03-13 04:11:12 @geofflangdale Fwiw, "lifetime emissions of five cars" metric is wrong &
2023-03-13 02:29:39 @osanseviero @draix @pau_martinez @HelenaCaseli @BrasileirasPLN @Fscrollini Thanks for the pointers! I was following some but not all!
2023-03-12 06:05:26 @geofflangdale @strubell @SashaMTL We make a similar magnitude argument in Section 5 of https://t.co/iHhFxGHg2p https://t.co/xfZeBMhl4I
2023-03-12 06:02:43 @marcslove @SashaMTL In follow-on work, we measured the energy use of ML at Google (inference, training, and research) to be <
2023-03-11 06:06:37 @rcopelan @DynamicWebPaige Chex and balances?
2023-03-10 19:34:10 @SashaMTL (To be very clear, none of the flaws were intentional, and we have discussed with the authors of the Strubell et al. paper. I do think making people aware of how that data is not actual measured emissions cost and is off by quite a bit is pretty important).
2023-03-10 19:12:06 @SashaMTL I definitely like the paper in that it focused attention on the potential issue and started a good (&
2023-03-10 08:35:47 RT @_jasonwei: Thanks @JeffDean for clarifying the Strubell+ '19 paper on energy consumption, which shocked me initially. - 5x overestimate…
2023-03-10 08:35:02 @sarahookr @Khipu_AI Maybe they're "rock star developers"? (of awesome AI talent at @Khipu_AI!)
2023-03-10 08:22:12 @luis_likes_math Are you frond of that look?
2023-03-10 08:17:47 @ThomasW423 @_jasonwei Indeed, it already has, and that particular NAS output (the Evolved Transformer) was open sourced for everyone to benefit from.
2023-03-10 06:11:20 Also highlighting a thread of my replies to @SashaMTL's original thread (both worth looking at if you're interested!). https://t.co/OcatqDPx9p
2023-03-10 06:08:07 An important thread about misrepresentation of CO2e numbers for different ML models. I believe it's good to use actual measurements of this data, rather than estimates that are flawed in a variety of ways. https://t.co/m1hfoU3TXv
2023-03-10 06:00:17 @MickeyShaughnes @SashaMTL All the infrastructure is accessible to external entities through our Google Cloud Platform. Table 1 in https://t.co/W0fNJGCV6c has columns that enables you to compare moving from avg US datacenter/electrical grid to Google datacenter, &
2023-03-10 05:45:14 @SashaMTL I have raised this exact point in the past with some Google authors, saying that it wasn't reasonable to use flawed estimates that were off by a factor of 3261X from the actual cost of training this language model, and that they should fix this before publishing. Sigh.
2023-03-10 05:37:45 @SashaMTL ... and off by ~118,000X if you consider training that language model on a Google TPUv2 in our Iowa datacenter. See Table 1 and Appendix C and D of https://t.co/W0fNJGCV6c for more details.
2023-03-10 05:32:14 @SashaMTL The actual emissions for training a language model of the scale examined by Strubell et al. is 3261X smaller than this flawed estimate even if you consider the "P100 in average US data center on average electrical grid" scenario they were evaluating ...
2023-03-10 05:22:25 @SashaMTL ... meaning the actual estimate for the one-time NAS search was off by 88X. This highly flawed estimate was then portayed as the emissions for training a language model, rather than the one-time cost of the NAS. ...
2023-03-10 05:19:23 @SashaMTL ...then estimated the CO2e as if it was run on P100 GPUs in average U.S. datacenter (much higher emissions, much worse PUE, so off by ~5X from actual running environment). Worse, they missed the use of smaller proxy tasks by So et al &
2023-03-10 05:12:50 @SashaMTL I completely agree with this take. Indeed, these issues have plagued other prominent work in this field. As one prominent example, the heavily cited Strubell et al. work took a one-time Neural Architecture Search done on TPUs in an efficient, low carbon Google Data center ...
2023-03-09 05:34:27 @jerome_massot @airfrance I didn't watch it. Just enjoyed the announcement..
2023-03-09 05:18:36 Somewhat related, I got a bit of a chuckle when I was flying Air France years ago, and they announced that the movie for the flight was going to be "Murder at 4 PM" (really "Murder at 1600", for 1600 Pennsylvania Avenue, aka the White House) https://t.co/LQnBt6bGQU
2023-03-09 04:48:25 Happy International Women's Day! Please celebrate the amazing women in your life and worldwide! https://t.co/VyX10VFrqb
2023-03-09 04:26:06 Open source software gives everyone the opportunity to help create breakthroughs in AI. At Google, we’re excited to collaborate on the OpenXLA Project to foster great AI tooling (great ML performance, framework compatibility, highly configurable &
2023-03-08 05:01:31 @DynamicWebPaige @jamescham Emacs. See relevant thread here with a few more details. https://t.co/yOQBhktu4Q
2023-03-07 07:25:03 @Scobleizer @GoogleAI It will likely improve over other speech systems (in general, systems with lower WER trained on variety of conditions/accents/acoustic settings are more robust). We also have Project Euphonia, which specifically targets diverse speech patterns. https://t.co/etFjleqL6a
2023-03-06 21:50:30 At @GoogleAI, we aspire to make sure that language and speech models work for everyone, by focusing on the world's 1,000 most spoken languages. Today we're describing our work on a Universal Speech Model that supports 300+ languages, described in the blog post below. https://t.co/PeCZFZdUR9
2023-03-05 22:36:32 Eligible @BlackInRobotics members, please consider applying for the Black in Robotics &
2023-03-05 10:00:00 CAFIAC FIX
2023-03-02 22:00:00 CAFIAC FIX
2023-02-27 01:00:00 CAFIAC FIX
2023-02-18 06:16:12 @roboticwrestler It does give one paws..
2023-02-18 06:13:13 RT @ChrSzegedy: Nice article in Quanta magazine on the prospect of combining automated formalization and theorem proving featuring @Yuhu_a…
2023-02-18 01:18:26 RT @alienelf: It's happening! I'm super excited to introduce y'all to @LelapaAI ~ AI for Africans, by Africans, solving African problems …
2023-02-18 01:18:12 @alienelf @LelapaAI @AtlanticaVC @mozilla @kbeguir I'm proud to be an investor and very excited to see how @LelapaAI grows and impacts Africa and the world! You all have an awesome initial team!
2023-02-16 22:05:01 RT @JeffDean: Part 5 in the series is out, discussing our work in robotics, especially allowing robots &
2023-02-16 21:52:18 Part 5 in the series is out, discussing our work in robotics, especially allowing robots &
2023-02-16 06:50:49 @SashaMTL @ayirpelle @rachelmetz Did you get them on sesame street?
2023-02-15 04:35:41 @mrogati Can you send me an email with the details and I can forward appropriately? Thanks.
2023-02-13 15:10:12 A very nice (long) slide deck, with each slide having a good visualization of an interesting trend. https://t.co/ECpbQ9pQOe
2023-02-12 01:03:20 RT @Rainmaker1973: Bonsai made with copper wire is a more and more traditional crafting technique for making impressive artworks. Pure copp…
2023-02-10 04:50:42 @dancherp Yes, fine to walk it. Might consider getting a small saddlebag and keeping a spare tube and tire levers plus a frame pump. Also if you're riding on roads with lots of debris, also consider tire liners. https://t.co/mvnIjJ7FlV
2023-02-07 23:34:56 RT @JeffDean: Part 4 in the series is out, by Sanjiv Kumar (on behalf of many!), covering a few different aspects of algorithms for efficie…
2023-02-07 20:48:50 Part 4 in the series is out, by Sanjiv Kumar (on behalf of many!), covering a few different aspects of algorithms for efficient deep learning. https://t.co/5bstN9Kf4w
2023-02-07 04:06:35 We're excited to share Bard soon. It utilizes a powerful version of LaMDA and is made possible by our foundational work on Transformers, dialog models, responsible AI, model safety techniques, and more. https://t.co/yKMA4d2OLo
2023-02-07 03:55:00 @AliciaCurth Thank you for your service to the community and to authors for investing the time and effort to write high quality reviews! This often goes underappreciated.
2023-02-05 02:14:14 A good but depressing visualization. https://t.co/HciL9XGUqL
2023-02-03 04:54:44 I love what @LastMileFund and @ruthef are doing to improve diversity in STEM. Although I've been trying to go metric, so I affectionately refer to them as the Last Kilometer Fund. https://t.co/DDVs8Tts0k
2023-02-03 01:52:32 @svpino It's a plot!
2023-02-03 00:06:00 RT @JeffDean: Part 3 in the series is out, by Mangpo Phothilimthana and @apaszke, covering the intersection of ML and computer systems (dis…
2023-02-03 00:05:40 Part 3 in the series is out, by Mangpo Phothilimthana and @apaszke, covering the intersection of ML and computer systems (distributed systems, compilers, hardware design, energy-efficient and low emissions computing, etc). https://t.co/oRNhgQ7OVV
2023-02-01 06:49:46 @sarahookr @ayirpelle Happy anniversary!
2023-01-30 01:00:00 CAFIAC FIX
2023-01-13 07:29:07 @indygupta @penberg @labarroso @IllinoisCS Wow, very nice and extensive reading list!
2023-01-13 06:19:12 @BeingNavadeep @penberg @labarroso Serengeti National Park in Tanzania, very near the Kenyan border
2023-01-13 03:24:07 @kvogt "We've reached cruising altitude. But please keep your seat belts fastened."
2023-01-13 03:11:35 @penberg @labarroso Glad you liked it! I present a visualization of long tails that didn't make it into the paper. https://t.co/g0ig7OBzuR
2023-01-13 03:09:48 Thanks, glad you liked "The Tail at Scale", @penberg! @labarroso and I had fun writing it. I'm attaching an image relevant to long tails that I usually use if I give a talk about this topic (always nice to be able to use ones' vacation photos!). https://t.co/01W5b4WVeY https://t.co/ElVgKU15Xk
2023-01-11 05:27:52 RT @nathanbenaich: TECHBIO BOOM! @BioNTech_Group to buy @instadeepai for £342M in cash + up to £200M more. In Feb ‘21, I wrote “I see…
2023-01-11 05:25:34 @nathanbenaich @ayirpelle @BioNTech_Group @instadeepai @kbeguir Congrats @kbeguir and everyone at @instadeepai! I still remember meeting many of you at @DeepIndaba in South Africa in 2018, and I've followed the company's impressive work and research since then. This is great to see, and a great success for the African AI ecosystem.
2023-01-10 15:06:54 Let the race for the quettaflops era begin! That quite-a-lotta flops. https://t.co/Tw91UxartB
2023-01-08 15:03:12 RT @culturaltutor: Bridges represent some of the greatest achievements of human engineering and architecture. Here are 12 of the most extra…
2023-01-08 06:19:32 @vitalygordon @DavidEpstein Indeed! I also liked his book "The Sports Gene". Thanks, @DavidEpstein!
2023-01-08 05:11:21 RT @deniswirtz: Here is our largest database of PhD fellowships yet: we found 195 fellowships and funding opportunities. For each fellowsh…
2023-01-08 05:06:36 This was a long but really good article: From the article: “This is my kryptonite, but this is her rocket fuel,” Jill says. “We’re like comic book superheros that are just as divergent as can be. I mean, her body has found a way around it somehow.” https://t.co/ogDePkMhl9
2023-01-06 07:44:32 @KLdivergence Congratulations!
2023-01-03 22:14:31 RT @sirbayes: I am delighted to announce that the "real" camera-ready version of my new book, "Probabilistic Machine Learning: Advanced Top…
2023-01-03 06:20:36 @Lan_Dao_ Ask any old crank how it works..
2023-01-03 04:21:38 @typedfemale Try not to let her get down. There are so many amazing institutions that will be great fits for her biochemistry &
2023-01-03 04:14:15 @Rainmaker1973 Knot fun.
2023-01-02 07:29:26 Happy New Year to all! May you all have health, happiness, and enthusiasm for everything great that you do!
2023-01-02 06:43:32 RT @DupuyTaylor: If I wanted to talk about the mathematics of board games in a course what should I include?
2023-01-02 06:42:24 @doctordave @DupuyTaylor Saw this comment and was curious. Found this page, which does a nice analysis. https://t.co/88amHSSIM1
2023-01-01 06:43:49 RT @PyTorch: If you installed PyTorch-nightly on Linux between Dec. 25 and Dec. 30, uninstall it and torchtriton immediately and use the la…
2022-12-31 03:38:57 @HowardKushlan I had an odd baseball career. Played in 3rd, 4th and 12th grade (did mostly soccer in between). Class A HS in Georgia, so not high level. I had decent accuracy and ok velocity. Happy to be 6-0 my senior year despite not playing for 8 years. Couldn't hit worth a darn, though!
2022-12-31 00:43:40 RT @rasbt: Vision transformers have taken the field of computer vision by storm, but what do vision transformers learn? ViTs have fewer in…
2022-12-30 22:55:22 As a former pitcher, I found this interesting.. https://t.co/DyL2WY67Ti
2022-12-29 05:49:47 @CatWorkers Cat, a tonic?
2022-12-29 02:59:52 @hieupham789 Congrats on the growing Phamily! Great picture!
2022-12-28 00:23:28 I highly recommend @vivnat's thread on the work! https://t.co/nGqgoz37Tj
2022-12-28 00:19:56 RT @vivnat: Delighted to share our new @GoogleHealth @GoogleAI @Deepmind paper at the intersection of LLMs + health. Our LLMs building on…
2022-12-28 00:10:14 RT @s_scardapane: *FlexiViT: One Model for All Patch Sizes* by @giffmana @Pavel_Izmailov @__kolesnikov__ @mcaron31 @skornblith @XiaohuaZhai…
2022-12-28 00:08:34 Some nice work by many @GoogleAI and @DeepMind researchers to assess the quality of answers to a variety of clinical QA datasets. In general, larger models+careful fine tuning on medical text performs better (many things to figure out before real use, but improving fast). https://t.co/qHrvyzgXR6 https://t.co/HrZPTxISaa
2022-12-27 18:39:41 @ankahira Had a lovely few days there years ago. It is indeed beautiful!
2022-12-25 04:00:28 @bretthmeyer @xriskology But effective...?
2022-12-22 02:51:02 @ankahira It's really nice to hear things like this.
2022-12-21 04:49:17 RT @GoogleAI: Following years of research featuring thousands of people who recorded over a million speech samples, we released Project Rel…
2022-12-18 22:46:23 RT @corl_conf: Aaaand the special innovation award goes to... https://t.co/54NJedntm8
2022-12-18 20:07:33 RT @mvechev: Tenure-track and tenured faculty jobs in AI/CS at INSAIT (https://t.co/iBXQuNuqwe), created in partnership with @ETH_en and @E…
2022-12-18 03:31:14 @TaliaRinger Oh no! I hope you make a full and speedy recovery!
2022-12-16 06:27:01 @EhudReiter I feel like it's fine to report the max (and min) as well as the mean and std. dev (or even the whole distribution).
2022-12-16 06:21:42 @TaliaRinger Couched in those terms, I'm all settee..
2022-12-14 06:50:26 RT @hausman_k: Introducing RT-1, a robotic model that can execute over 700 instructions in the real world at 97% success rate! Generalize…
2022-12-10 23:09:47 Nice to see the relatively steep slopes on wind and solar. Wish the negative slopes on coal and gas were more vertical. https://t.co/VuMzBCj1du
2022-12-10 23:04:12 @emilymbender I completely agree. Accuracy in one's scientific statements is very important.
2022-12-09 06:30:41 @_ejgonz_ @Google @twi_mar Welcome to Google!
2022-12-08 13:00:00 CAFIAC FIX
2022-12-07 08:00:00 CAFIAC FIX
2022-11-13 03:28:06 @DynamicWebPaige Welcome!
2022-11-11 15:47:58 @A__Diack @eyram_yvonne Nice!
2022-11-10 19:41:13 RT @GoogleAI: As language models increase in scale they sometimes exhibit various useful emergent capabilities. In today’s post, we explore…
2022-11-10 04:13:18 RT @jekbradbury: Can multi-100B param language models be served efficiently? We think so! Today we’re announcing the PaLM inference paper a…
2022-11-09 20:09:26 Some really great visualizations and demonstrations by @adamrpearce and @nicki_mitch around the key ideas behind Federated Learning, and how it enables learning communally-useful ML models without ever bringing all the raw data together. https://t.co/poamaT8ZWD
2022-11-09 05:38:05 @IllinoisCS @intel @uofigrainger @UofIllinois @tianyin_xu Congrats, @tianyin_xu!
2022-11-08 06:44:48 Accelerating climate action at Google and beyond.. https://t.co/vHQEXqNDdE
2022-11-07 22:34:03 RT @GoogleAI: Read about our latest work from a research effort called Infinite Nature, which can produce high-quality flythroughs of richl…
2022-11-06 19:51:52 @kelseyhightower I love Barcelona. Thanks for the view from the high tower!
2022-11-05 22:36:40 @mcmillen @laurenweinstein Should have used name hashing for ordering, with different salt every day.
2022-11-05 03:27:03 @mariana057 The pain is real.
2022-11-04 23:26:23 RT @sundarpichai: 3/ Using new advances in AI, we're bringing flood alert coverage to 18 more countries. And we're announcing Google FloodH…
2022-11-04 23:26:13 RT @sundarpichai: 2/ Our wildfire tracking system will now identify and track wildfires in real time, helping predict how they will evolve…
2022-11-04 23:26:08 RT @sundarpichai: 1/ From today's AI@ event: we announced our Imagen text-to-image model is coming soon to AI Test Kitchen. And for the 1st…
2022-11-04 23:24:29 To coincide with our AI@ event this week, I wrote up a brief blog post, touching on three important topics:Supporting 1,000 languages with AIEmpowering creators and artists with AIAddressing climate change &
2022-11-04 19:21:06 @dave_andersen Congrats on the award! This is fantastic recognition of your work and that of your students! We are all weiser for your work.
2022-11-04 19:10:34 @the_transit_guy Sounds like an origami flex.
2022-11-04 19:08:37 @ankahira Congratulations!
2022-11-04 03:04:40 @sarahookr Encounters with the Archdruid, by John McPheehttps://t.co/jj4zpxnTfhThe Wild Trees, by Richard Prestonhttps://t.co/MC8qFMYRcmCadillac Desert: The American West and Its Disappearing Water, by Marc Reisner. https://t.co/RzrQ3GvkNjAny of this series: https://t.co/4aZJk9AaHN
2022-11-03 05:38:37 @thiagoghisi @GergelyOrosz @vinthanedar A couple days per week, most of the day.
2022-11-03 05:37:27 "1 mile setbacks from turbines to property lines" (i.e. 1 turbine on a 2 mile by 2 mile property?!)"A 35 dBA acoustical limit 24/7 at property lines" is completely absurd, given that 40 dB is equivalent to "quiet library sounds". https://t.co/Y6zaoIrybd
2022-11-03 05:21:17 RT @curious_founder: By far the most unique tactic, I saw in Droz' training materials was his suggestion that clean energy opponents never…
2022-11-03 05:07:03 Ugh. This thread is an annoying account of tactics used to prevent wind and solar projects. https://t.co/y4zPsrXSQh
2022-11-03 04:56:57 RT @caesars: Folks - I’m thrilled to introduce you all to @arta_financeWe started Arta because we believe financial superpowers shouldn’t…
2022-11-03 03:30:43 @thiagoghisi @GergelyOrosz @vinthanedar We still do. We're working together tomorrow.
2022-11-02 13:53:51 Our AI@ event is happening now! Watch live at youTube link below! https://t.co/w0whL3E5gX
2022-11-01 00:10:20 @mariana057 Maybe you need a comforter upon hearing this.
2022-10-30 23:46:01 RT @highbrow_nobrow: Accidental philosophy: “Please pay your parking fee before existing.” https://t.co/ECoUw5yX0e
2022-10-29 20:03:42 RT @YouAreLobbyLud: This seems amazing. Belgium now requires, BY LAW:1) public indoor spaces must display CO2 level2) risk analysis and…
2022-10-29 19:58:32 @moultano I am here for aleph this..
2022-10-29 19:53:26 @Anon__08 @jsomers @NewYorker Nope..
2022-10-28 23:11:20 @kelseyhightower Congratulations, Kelsey! You've more than Distinguished yourself over the years!
2022-10-28 20:29:39 I'm excited to see our @GoogleAI researchers share how Google’s AI technologies can help people, particularly to stay safe, better connect, and be creatively inspired.Join us on Nov 2nd to watch AI@ live! https://t.co/ug7ac8GDTk
2022-10-28 03:12:36 @balajiworld No embarrassment! Different opportunities present themselves over time, and the important thing is to be accepting of others and to be curious when you do get the chance to travel.
2022-10-28 03:07:57 @minhtrann @dave_andersen First and second grade, near Little Rock. We lived on five acres that was a former chinchilla ranch (parents remodeled main building into a house), &
2022-10-27 11:40:06 @lamechthinkbig @dave_andersen I wish there was (I'm not involved in the project, but there's a GitHub link for it)
2022-10-27 02:08:55 @jbrownkramer @dave_andersen I guess that's a goal in the search of the state space.
2022-10-27 02:04:15 @jbrownkramer @dave_andersen Sorry! I actually want to visit all the states I haven't been to!
2022-10-27 02:01:52 @lk_bell @dave_andersen I did it as "stayed == spent at least one night there"
2022-10-27 02:01:08 https://t.co/OFvpBZQUgl
2022-10-27 01:57:59 (Feel free to reply with yours! There's also an EU version)
2022-10-27 01:50:42 RT @GoogleAI: On Nov 2 at 9am EST, we're excited to host our first AI@ event: a showcase of our latest and greatest work in AI. Tune in liv…
2022-10-27 01:45:42 This was kind of fun (although like @dave_andersen, I suspect I'm forgetting a few road trips or short stops here and there).https://t.co/JsFkJzWEUZ https://t.co/JKnAPP4zDl
2022-10-26 14:11:08 @moyix @danrkports You can use net, but not use nets.
2022-10-25 21:46:52 RT @GoogleAI: Learn how Open Images V7 expands the Open Images dataset with new point-level label annotations, which provide localization i…
2022-10-25 03:27:43 @wowinteresting8 Please power cycle your phone
2022-10-23 21:34:41 RT @quocleix: New open-source language model from Google AI: Flan-T5 Flan-T5 is instruction-finetuned on 1,800+ language tasks, leading…
2022-10-23 17:25:01 If you're at #ECCV2022 (virtually or physically), @GoogleAI researchers that are also attending would be happy to chat with you (about their work, your work, career advice, and many other topics!). See below for list of 60+ accepted Google-affiliated papers + workshops/tutorials. https://t.co/3FRcQsTadv
2022-10-22 21:44:11 A recent blog post describing some of the @TAE/@GoogleAI collaboration.https://t.co/piVYFpQgPk
2022-10-22 20:29:08 Nice article about the effort to develop fusion energy, focused on @TAE. Amazing for the world if we can do it! @GoogleAI has a many-year collaboration w/TAE to use machine learning to aid in creating stable fusion reactions.https://t.co/JDyFXORKT7
2022-10-22 20:18:40 RT @sundarpichai: Work from our @GoogleAI team - training robots to play table tennis to better understand how they learn in dynamic + high…
2022-10-22 12:41:33 @RisingSayak @ksoonson I'm glad you liked the work!
2022-10-21 22:13:13 @dave_andersen I'm willing to have the "In my day, when I was your age, my acoustic coupler topped out at 110 baud" discussion, if helpful.
2022-10-21 21:31:32 RT @BrentToderian: Decades ago, many of us urbanists had this image in our slide deck to illustrate how simple exercise has been designed o…
2022-10-21 02:46:28 @bonadossou @mcgillu @McGill_Eng @McGill_CIM Congratulations!
2022-10-20 20:11:40 RT @rikelhood: The recordings from the Google Workshop on Sparsity and Adaptive Computation are now available at:Day 1: https://t.co/8yUH…
2022-10-20 13:00:42 RT @kelvin_guu: New from Google Research! Language models perform amazing feats, but often still "hallucinate" unsupported content. Our mod…
2022-10-19 18:43:00 RT @JeffDean: @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. Fo…
2022-10-19 17:47:57 @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. For some reason it wasn't showing up on this list of articles by the author, but I found it at this URL:https://t.co/pkAQ0IRmBj
2022-10-19 11:30:13 RT @GoogleAI: Applications are open for CSRMP, a mentorship program that matches students from historically marginalized groups in computin…
2022-10-19 11:29:49 RT @GoogleAI: Learn about AudioLM, an audio generation framework that demonstrates long-term consistency (e.g., syntax in speech &
2022-10-19 11:29:40 RT @GoogleAI: Image captioning datasets can be limited for languages other than English. So, today we are releasing the Crossmodal 3600 dat…
2022-10-19 11:07:27 RT @GoogleAI: Check out a new table tennis platform built to study problems that arise from robotic learning in a multi-player, dynamic, an…
2022-10-19 04:40:32 RT @BarackObama: On this International Day of the Girl, we stand in solidarity with the courageous Iranian women and girls who have inspire…
2022-10-18 05:47:28 @docmilanfar Do not begrudge them. The koala tweet of mercy is not strained.
2022-10-16 04:23:13 @__apf__ That's a bit disarming.
2022-10-15 16:59:15 @dataandpolitics This is so hateful and wrong. . I'm sorry you have to deal with this while also grieving for your loss. Please take care of yourself!
2022-10-15 03:25:28 RT @stshank: Is the Pixel 7 Pro as good as a "real" camera? With @madebygoogle bragging of its "pro-level zoom" and AI smarts, I wanted to…
2022-10-14 16:07:47 RT @ZoubinGhahrama1: @GoogleAI and @emblebi release 40 million annotations of previously uncharacterized proteins in the UniProt database!…
2022-10-14 06:07:54 RT @NaqiyFL: Transit is way more safer than driving, countless lives would be saved if we invested in transit as much as new roads https://…
2022-10-13 01:28:56 RT @arankomatsuzaki: Mind's Eye: Grounded Language Model Reasoning through SimulationImproves reasoning ability using MuJoCo simulations…
2022-10-12 22:41:07 RT @Google: When governments and aid organizations are able to understand the locations and geometry of buildings, they’re better able to u…
2022-10-12 20:55:30 Nice article about our TPUv4 system.https://t.co/7AsotkcNw2
2022-10-12 20:54:14 RT @Google: Google execs, @JeffDean and Yossi Matias (@ymatias), discuss the potential of AI in Africa and our commitment to invest in rese…
2022-10-12 15:41:28 @tejuafonja @lautechofficial @Saar_Uni Congratulations!!
2022-10-12 01:21:15 RT @AmyCChou: Check out our latest work on AI for fetal ultrasound in under-resourced settings. The models run on-device, without internet…
2022-10-11 06:21:51 RT @docmilanfar: Never miss an opportunity to look younger and more fancy than your co-authors https://t.co/kkdBvvHfrD
2022-10-10 13:35:21 RT @A2_SV: What a rewarding feeling it is to be able to find three A2SVians in Palo Alto, the heart of Silicon Valley, pursuing their caree…
2022-10-10 13:35:09 @A2_SV It was great to chat with the four of you!
2022-10-08 23:45:29 @TaliaRinger Finland and I think Switzerland do this, and it makes total sense to me.https://t.co/aWObhpD43C
2022-10-08 04:51:07 A towerful demonstration of the zoom capability of the new #Pixel7Pro https://t.co/E3w3VucGfS
2022-10-08 04:30:53 RT @CSforALL: We are delighted to have @1_d_sanders as a guest speaker at this year's #CSforALLSummit!With more than ten years of experti…
2022-10-08 01:11:03 Could we get a couple of these working on El Camino in the Mountain View, Palo Alto, Menlo Park stretch, please? https://t.co/ib2PwD1ZGl
2022-10-07 04:11:41 @pmclsf I am so sorry to hear this. My deepest condolences to you and all your and your husband's family and friends.
2022-10-07 02:16:39 @judyfhoffman Congrats on the well-deserved recognition of you and your students' work!
2022-10-07 02:14:20 RT @madebygoogle: Bye bye, blur.#Pixel7 and #GoogleTensor G2 help ensure faces are incredibly sharp by automatically fusing images from b…
2022-10-07 02:13:21 RT @madebygoogle: Taking amazing photos in the dark just got even easier with #NightSight on #Pixel7 #GoogleTensor G2 uses machine lear…
2022-10-07 01:17:05 RT @docmilanfar: Today we announced a new feature on Pixel 7/Pro and @GooglePhotos called "Unblur". It's the culmination of a year of inten…
2022-10-06 17:53:52 RT @sundarpichai: Excited to launch Pixel 7 and Pixel 7 Pro with our next generation Google Tensor G2 chip, which brings state-of-the-art #…
2022-10-06 14:31:05 RT @BrentToderian: “Between 2021 and 2026, Paris will gain 180 kilometers (112 miles) of new permanent segregated bike lanes. As part of th…
2022-10-05 08:46:57 Bleh. @UofIllinois. <
2022-10-05 02:07:32 @takriti @Google @amazon @Microsoft @Meta @Apple @uiuc @UofIllinois Oops, yes. Sorry.
2022-10-05 01:35:28 Nice to see @Google, @Amazon, @Microsoft, @Meta, @Apple, and @uiuc all collaborating to improve speech recognition for those with diseases that affect speech such as ALS, Parkinson's, cerebral palsy, Down syndrome and other diseases.https://t.co/e5kSKLRwpI
2022-10-05 00:59:31 @Gabriele_Corno Wow! Was that 7 rotations?
2022-10-05 00:53:12 New work from @GoogleAI and many collaborating institutions on the use of deep learning for histologic grading of great cancer and assessing prognosis. Congrats to all involved! https://t.co/PhWIFpe7t6
2022-10-01 11:52:41 RT @A__Diack: https://t.co/Fb9XtL2rAD is committing $25M to fund solutions that accelerate progress towards the Global Goals.https://t.co/…
2022-10-01 00:17:27 RT @NanaYaaSally: Please don’t forget to apply for the @black_in_ai funding for attending this year’s workshop co-located with @NeurIPSConf…
2022-10-01 00:08:58 @dave_andersen Is that an N95 hat?
2022-09-30 04:13:54 RT @poolio: Happy to announce DreamFusion, our new method for Text-to-3D!https://t.co/4xI2VHcoQWWe optimize a NeRF from scratch using a…
2022-09-28 06:01:03 Happy 24th Birthday, Google! https://t.co/8TK1wbCoSL
2022-09-26 05:10:57 RT @BrentToderian: The Dutch invest €595 million annually on urban biking, resulting in €19 BILLION saved in public health care costs alone…
2022-09-26 00:24:44 RT @NaturePortfolio: From connectomics to behavioural biology, artificial intelligence is making it faster and easier to extract informatio…
2022-09-24 03:22:35 @ShriramKMurthi That's how he roles..
2022-09-23 18:50:48 @vncsdb @stardazed0 @GoogleAI On the margin, this seems like an improvement..
2022-09-23 03:58:14 RT @susan_dagostino: The next generation of mathematicians and computer scientists #HLF22 @HLForum https://t.co/595uAg9eWa
2022-09-23 03:50:52 TensorStore offers pretty nice read and write throughput and has good scaling properties. https://t.co/lqhbhJEfaW
2022-09-23 00:16:28 Infrastructure that underlies great research often doesn't get the attention that it deserves. This post discusses TensorStore, a (now) open-sourced high-performance, scalable array storage system. It enables @GoogleAI's connectomics work, LLMs, + more...https://t.co/9GpIByD4Oq
2022-09-20 23:39:38 RT @KanubalaD: As the 9th @HLForum is currently ongoing, this is probably a good time to share some old photos I took with a few laureates…
2022-09-20 23:36:40 @KanubalaD @HLForum Thanks for sharing! It was great chatting with you at the #6thHLF!
2022-09-20 04:40:23 Nice work, @chelseabfinn! I love these videos of explaining complex topics to people at different levels, and you did a really nice job in this set of explanations!Full video here (instead of teaser link in the original tweet):https://t.co/X6J5hvEnjW https://t.co/czQOwGTO8x
2022-09-18 04:55:19 @tejuafonja @HLForum Have fun! Wish I was able to attend this year, but unfortunately, I won't be able to make it.
2022-09-17 03:51:11 RT @wowinteresting8: Best life hacks for you https://t.co/OSyeqRyHVv
2022-09-16 05:28:13 As we train new and interesting multimodal models of various kinds, it's important that they work well in many languages. PaLI is a model that can perform a variety of language+image understanding tasks in 100 languages. https://t.co/PMzoGBwTs8 https://t.co/EMlEi1YwAy
2022-09-12 16:38:46 RT @kaggle: Take part in our 6th annual Machine Learning and Data Science Survey! Think Stack Overflow's Developer Survey but for data sci…
2022-09-09 22:09:16 I got to hear earlier versions of this work and it really does have amazing sound quality, given a very small amount of conditioning audio. Listen to the couple of samples in the tweet below! https://t.co/WCaR0PfNxO
2022-09-08 03:05:36 RT @LiamFedus: Our survey on sparse expert models describes the advances over the last decade, discusses some difficulties, and presents ou…
2022-09-07 05:54:26 @DimitrisPapail Very nice and great directions on this research work. Congrats, @shashank_r12!
2022-09-07 05:40:33 This project nose the "sweet smell of success"... Or maybe basil with hints of green tea and mango chutney. The kind of things you might experience daily on the ol' factory floor. https://t.co/7TkBfS5hvx
2022-09-05 16:14:31 @Gabriele_Corno If you crane your neck, you can see some nice castles in the background.
2022-09-03 15:44:27 RT @Google: A big congrats to 2022’s cohort of Google PhD Fellows, who are part of our program that recognizes and supports graduate studen…
2022-09-02 18:50:59 RT @A__Diack: Congratulations to all Google PhD Fellowship 2022 recipients. Recipients are provided with funding for research activities…
2022-08-30 13:57:55 @TaliaRinger Aleph them?
2022-08-28 00:45:20 @mariana057 @MuellerSheWrote @DannyDeraney @natsechobbyist @christoq @steveking_ @happycamper281 @GadflyDer @MarciaPopp1 @JohnFetterman @rmayemsinger @lychpsych @MezaroDana @TeaPainUSA @Darth_Fugue @MobBossTrump @WhippetRun @TheUSASingers @Wieneraaron @SeanMastersons I am honored. Though I hope these are not the circles of Dante Alighieri.
2022-08-27 01:49:40 @TansuYegen @gpetzinger I spot... Mini Driver?
2022-08-26 01:27:50 @dataandpolitics https://t.co/0UY6RYzfOm has some postings that look relevant. Various keyword searches for things like[ Firmware principal ][ Performance principal ]turn up some possibilities that could work.
2022-08-25 18:51:45 Great to see work to combine the best aspects of #Imagen and #Parti into a hybrid system with even better capabilities. https://t.co/VVEUi34FwI
2022-08-25 18:50:27 RT @jasonbaldridge: Exciting news: #Parti and #Imagen teamed up to create a hybrid system with Parti creating 256x256 images which then rec…
2022-08-20 05:41:42 @mariana057 A real spring in her step!
2022-08-20 04:05:57 I also follow @ZLabe and find his visualizations are really great at highlighting very troubling climate trends.@moultano does a nice job of explaining why this particular one is so visceral and compelling. https://t.co/E8oOQyUxw9
2022-08-17 13:50:07 RT @hausman_k: We have some exciting updates to SayCan! Together with the updated paper, we're adding new resources to learn more about thi…
2022-08-17 06:28:21 RT @jazz_inmypants: I cannot accept that a human being is capable of this https://t.co/RSvcPvOnzC
2022-08-17 06:03:19 Watch the video to see how @GoogleResearch and Everyday Robots are working together to combine language models and robots to improve robotic learning and the ability to get robots to do real-world tasks. https://t.co/ZscU4M2QG2
2022-08-16 05:20:18 @DynamicWebPaige Things are getting... spurious?
2022-08-15 05:14:16 @satnam6502 I now pronounce you... ... ...
2022-08-15 03:20:56 @WUTangKids @TheWienerCircle Dog-gone-it, no longer dog-on-it.
2022-08-14 18:36:33 @AnimaAnandkumar Oh no. Not the best way to spend a birthday . I hope you have a speedy and full recovery.
2022-08-12 19:46:58 Nice article (in German) about the rollout of climate-friendly and fuel-efficient routing rolling out in Google Maps in more locales around the world!@googleDE https://t.co/KCon2Qlb4V
2022-08-11 18:12:15 @the_transit_guy Take the plunge!
2022-08-07 22:58:09 I will sheepishly retweet this.. https://t.co/ptD0eg39H6
2022-08-07 04:25:29 @johnregehr @danrkports @schemeprincess Unless there's been some sort of caper caper.
2022-08-07 04:07:05 @Diyi_Yang @Stanford Congratulations! Stanford is lucky to have you!
2022-08-06 23:39:04 RT @QuantaMagazine: Scientists are making rapid progress in mapping neural connections within the brains of worms, flies, mice and even hum…
2022-08-06 21:22:25 @hbhammel Kalalau Trail is a reasonably strenuous hike but worth doing if it makes sense (you don't need overnight permits to go to first beach).Koloa Zipline was fun.Waimea Canyon is great.Lots of great beaches (I liked Kalihiwai Beach for swimming/bobbing in moderately big waves)
2022-08-05 05:39:43 @chelseabfinn @TianheYu @DorsaSadigh @svlevine @StefanoErmon Congrats, @TianheYu, and congrats to the entire IRIS lab on the first PhD graduate!
2022-08-04 04:03:28 RT @wowinteresting8: Comparing the speed of light https://t.co/LxdZ1rkehO
2022-08-03 02:29:40 @nakadori @black_in_ai @FrontiersIn @DMachuve @ezinne_nwa @NM_AIST @UCBerkeley @UdsmOfficial Congratulations!(I hope when you run inference for a young patient that it's called a "chick up"!)
2022-08-02 02:41:37 Some nice work in getting NeRF models to run effectively at high frame rates on mobile phone hardware. Could lead to really natural on-device experiences (watch video!).Project led by @GoogleAI intern @ZhiqinChen3 along w/ Thomas Funkhauser, @PeterHedman3, &
2022-08-01 03:07:33 @TaliaRinger Happy birthday!
2022-07-31 17:10:38 RT @wowinteresting8: Sculptures from chocolate https://t.co/261W7uwwOL
2022-07-29 23:37:19 RT @sirbayes: I am delighted to announce that the camera-ready version of my new book, "Machine Learning: Advanced Topics", is finally avai…
2022-07-28 14:10:34 @TaliaRinger I would have expected pudding..
2022-07-28 05:44:54 @__apf__ You've never Felt better!
2022-07-28 05:44:16 RT @sundarpichai: Sharing a few @googlemaps updates today, including more granular cycling route info like steep hills, major roads vs loca…
2022-07-28 01:52:46 RT @simi_97k: Come join us at #DecodewithGoogle 2022 where @ManishGuptaMG1 and I will be sharing about how our Research team is tackling un…
2022-07-27 04:33:28 I was honored to attend Khipu in 2019 (also in Montevideo), and it was a wonderful experience. I met so many amazing people and heard many interesting talks! If you're interested in machine learning research, you should consider attending the 2023 edition! https://t.co/pfqYg7d5Pe
2022-07-27 03:17:42 @AmyCChou Glad to hear your surgery and care went well and wishing you a speedy recovery!
2022-07-27 02:52:55 This is the result of a great collaboration between the Google Brain team in @GoogleResearch and the Google Core Developer Tools team.
2022-07-27 02:49:17 ML assisted coding is helping thousands of Google software engineers."We see a 6% reduction in coding iteration time (time between builds and tests) [vs. control]" &
2022-07-27 00:21:43 RT @tkasasagi: Today I will give the opening keynote for the Digital Humanities Conference 2022. I would like express my sincere gratitude…
2022-07-25 04:40:16 @heiga_zen Congrats!
2022-07-25 04:29:18 RT @kat_heller: I was on the award committee. The committee was an incredible group of knowledgeable researchers who looked for impactful f…
2022-07-25 03:10:11 @LotfiSanae Congrats!
2022-07-25 03:09:44 RT @LotfiSanae: This Moroccan Arab Muslim first-generation woman just gave her first long talk for her award-winning paper at #ICML2022! I…
2022-07-25 03:00:16 A really good thread about the importance of welcoming new researchers to a field. Best paper awards are a way of recognizing work that helps us think about problems from new perspectives &
2022-07-25 00:36:18 @lingjuan_lyu Congratulations!
2022-07-25 00:11:34 RT @natashajaques: Incredibly honoured to have received an honourable mention for best paper at ICML this year https://t.co/5dtPtP3j0W
2022-07-25 00:11:18 RT @hima_lakkaraju: Super excited that our paper on "Which Explanation Method Should I Choose? A Function Approximation Perspective to Char…
2022-07-24 21:29:20 @ch3njus Thank you!
2022-07-23 04:10:31 RT @dmdohan: Happy to release our work on Language Model Cascades. Read on to learn how we can unify existing methods for interacting model…
2022-07-22 23:50:11 @DynamicWebPaige You should give it a shot!
2022-07-22 04:22:39 Wow."Even before this year’s ...announcement, the Institut des Hautes Etudes Scientifiques... boasted a remarkable statistic. Since its founding in 1958, the institute has had 12 permanent mathematics professors
2022-07-22 04:11:32 @ericjang11 Optimality is all you need sounds like the best title.
2022-07-21 06:13:43 @JamesLarus @icmlconf @quocleix @MarcRanzato @rajatmonga @greg_corrado @AndrewYNg Winner was "Poisoning Attacks Against Support Vector Machines". Congrats to Battista Biggio, Blaine Nelson, &
2022-07-21 02:22:02 The paper is affectionately known as the "cat neuron" paper. This is in preference to the "creepy face guy neuron" paper. https://t.co/CJllA9yIab
2022-07-21 02:12:01 One decision we made was to use local receptive fields, but not to tie the weights at different spatial positions, since this seemed more biologically plausible. In retrospect, tied weights (e.g. convolutions) would have been more data efficient, probably. https://t.co/CkbjZjlHyP
2022-07-21 02:11:09 This was a quite large model for the time, with ~1B trainable parameters.
2022-07-21 02:07:07 John @markoff wrote a nice NY Times article about the work:https://t.co/6bR3saoazF
2022-07-21 02:05:31 This was one of the first uses of DistBelief, the large-scale neural network training system we built that was a predecessor to TensorFlow.Paper: https://t.co/CrPzzrDA4h
2022-07-21 02:02:23 Honored that our 2012 paper "Building High-level Features Using Large Scale Unsupervised Learning" received an @icmlconf Test of Time Award honorable mention!Joint work with @quocleix, @MarcRanzato, @RajatMonga, Matthieu Devin, Kai Chen, @greg_corrado, myself, &
2022-07-20 03:49:51 RT @GoogleAI: Attending #ICML2022? Drop by the Google booth and chat with researchers to learn about our exciting #ML research that spans t…
2022-07-19 04:37:46 RT @RESCUEorg: ‘Our warning’: IRC teams report that people have already started dying of hunger in Ethiopia, Somalia, Kenya. Over 3 mill…
2022-07-19 04:10:46 RT @simongerman600: Chicago mapped only by two years worth of traffic crash data. No background map, just crashes. Source: https://t.co/9il…
2022-07-17 05:56:14 @giffmana @hardmaru @tkasasagi Distillation..
2022-07-17 05:52:07 @QAGreenways I'm so confused by the arrows on the ground..
2022-07-17 05:30:37 @docmilanfar I've had this. It's good!
2022-07-16 23:41:19 I knew about many of these but not all. I have a couple new kinds that sound good to try! https://t.co/BBaHKd5x0V
2022-07-16 18:58:15 @ghouqueue @nextdoorsv You are prophetic, @ghouqueue! From today's free Palo Alto paper. Seven stories! What's next?!!? Eight?!? https://t.co/OxnSN7wYzd
2022-07-16 05:15:41 @_arohan_ The Ngrammerians would like a few words..
2022-07-16 05:02:36 RT @_LXAI: Are you ready? LXAI at #ICML2022 You can get your tickets here https://t.co/WUu1W2rPNB
2022-07-16 05:00:37 This is great: a short, easily memorable for those in distress. 988. https://t.co/GXBp1bqTE2
2022-07-15 05:51:45 RT @_LXAI: Are your bags packed and ready to go? Come and join us in Baltimore for the LatinX in AI at #ICML2022. If you haven't registered…
2022-07-15 02:57:54 @DynamicWebPaige @SFJAZZ Steadily getting packed means Glass is half full?
2022-07-15 00:38:23 RT @dustinvtran: In our work "Plex", we propose a framework for reliability in AI. We also introduce new models (ViT-Plex &
2022-07-14 05:07:10 The barista should take pride in this art! https://t.co/kONxF27GSc
2022-07-14 05:01:41 @rao2z It's a walking program, going hither and yann while descending gradients.
2022-07-14 03:16:59 RT @ZoubinGhahrama1: Excited to announce that we've set up the first Google Research Hub in Australia! We've got a great leadership team, w…
2022-07-13 18:30:57 RT @hausman_k: Have you ever “heard” yourself talk in your head? Turns out it's a useful tool for robots too! Introducing Inner Monologue…
2022-07-12 04:02:12 RT @CompArchSA: Just announced: Our Summer DEI Reading Group for 2022 begins next week on July 13th at 11 AM ET with a welcoming address by…
2022-07-12 03:39:58 RT @marinakoren: There are so many galaxies in here. So, so many galaxies. https://t.co/3pJ8hXGMhM https://t.co/3nbNladmjy
2022-07-12 03:20:42 RT @ISSAboveYou: You've probably heard the @Space_Station orbits the earth 16x every day right? So that means the astronauts can see 16 sun…
2022-07-12 03:16:40 Exciting new work combining pre-trained models to enable robots to follow natural language instructions to accomplish navigational goals (zero-shot!).Work by @shahdhruv_, @blazejosinski, @brian_ichter &
2022-07-12 02:39:58 The universe is so vast."Webb’s image covers a patch of sky approximately the size of a grain of sand held at arm’s length by someone on the ground – and reveals thousands of galaxies in a tiny sliver of vast universe."See: https://t.co/k851kPo8pb https://t.co/e9DAu7raUs
2022-07-11 14:04:53 RT @KarakeMark: Going to @TEDTalks https://t.co/1Lu6bJL31k
2022-07-11 03:24:51 @karinv @xamat My dad the doctor used to refer to this as "occulogastric disproportion".
2022-07-09 11:53:05 RT @QueerinAI: Heads up to all NLP researchers Beside our workshop, Queer in AI is organizing *two* socials at @naaclmeeting #NAACL202…
2022-07-09 11:52:17 @lifesafeast @hotel_diderot Seriously, though, that hotel and jam armoire (a jamoire?) look delightful.
2022-07-09 11:43:33 @lifesafeast @hotel_diderot Not my jam...
2022-07-09 11:36:33 I'm just tired https://t.co/EMFWeSOWyI
2022-07-09 11:24:03 @GittaKutyniok @researchbavaria @LMU_Muenchen Congratulations!
2022-07-08 21:50:18 I enjoyed this rainbow sprinkler this morning https://t.co/1ueqDfTEUp
2022-07-06 21:30:13 RT @DynamicWebPaige: If your academic research in computing and technology addresses the needs of historically marginalized groups,…
2022-07-06 21:29:12 RT @WiMLworkshop: 2/2 Learn more and apply by July 13, 2022 at https://t.co/1NSYlm0tXk
2022-07-06 21:29:05 RT @WiMLworkshop: 1/2 Sharing an announcement from WiML corporate partner, @Google Applications are open for AIR, a program that recogni…
2022-07-06 21:22:18 RT @MarksLarks: Life cycle of the blackberry... https://t.co/GePemAhurB
2022-07-06 21:21:48 RT @acarroll_ATG: Release of DeepConsensus v0.3. ~5x faster than v0.2. Improved accuracy (yield of Q30 reads on a 24kb dataset increased to…
2022-07-06 21:21:23 RT @orevaahia: Attending the @DeepIndaba in Tunisia? We are accepting submissions for spotlight talks at the ML Efficiency workshop. We are…
2022-07-06 21:21:02 RT @sarahookr: It is official! Our "Broadening Research Collaborations" workshop has been accepted to @NeurIPSConf 2022! w incredible c…
2022-07-06 21:07:21 Improving things like compiler inlining decisions is something that is near and dear to my heart (one of my first papers was "Towards better inlining decisions using inlining trials"). Now it's great to see this problem tackled with reinforcement learning to great effect! https://t.co/thz9slrQOd
2022-07-06 14:55:33 RT @googledevjp: Google # Google Developers Machine Learning Bootcamp https://t.co/NGt757GWv7…
2022-07-06 14:55:21 This is a great opportunity for anyone interested in acquiring ML developer skills and experience. Apply by July 16th! https://t.co/4gJoWgyaBq
2022-07-05 21:51:11 RT @A__Diack: If you have been admitted to the Deep Learning Indaba Tunisia conference but need travel support please apply here: https://t…
2022-07-03 13:35:09 RT @emollick: The public colleges of New York, Texas, and California are great engines of social mobility in America. They move many born i…
2022-07-02 09:47:54 RT @the_transit_guy: By 2035, Amtrak plans to add train service between Detroit and Toronto with a 4:46 travel time. In China, this would t…
2022-07-02 09:30:28 RT @SwissEmbassyUSA: Marriage for all! Nine months after a nationwide referendum, same-sex marriage is now legal in #Switzerland as…
2022-07-02 09:24:11 RT @gunsnrosesgirl3: Sea lions riding huge waves near Santa Barbara Island, California filmed by Ryan Lawler of Pacific Offshore Expediti…
2022-07-01 19:49:05 If you are a member of a historically underrepresented group in computer science and are interested in conducting computer science research, the CSRMP matches you in small groups with a Google mentor to help you conduct research &
2022-06-30 20:52:18 If you're interested in the intersection of ML and computer systems, you should definitely consider attending #mlsys22! Registration is open now for the conference in Santa Clara, CA (conference dates: Aug 29th through Sep 1st, 2022) https://t.co/l63mZad67h
2022-06-30 20:17:45 RT @bneyshabur: Very excited to announce a significant milestone in expanding reasoning capabilities of language models! We introduce #…
2022-06-30 17:30:56 RT @alewkowycz: Very excited to present Minerva: a language model capable of solving mathematical questions using step-by-step natural lan…
2022-06-30 16:42:49 @alienelf Congrats!
2022-06-30 14:49:20 @wowinteresting8 Dam, that's interesting.
2022-06-30 09:43:50 RT @CoderMiddy: This deck of cards has a bar code printed on the edge of each card. Scanning these bar codes would reveal where every card…
2022-06-29 21:47:14 Yikes. I don't think it's supposed to be 87 degrees F at that latitude, even in June.. https://t.co/IRuna7EFsB
2022-06-29 21:39:11 @shakir_za @siminyu_kat @techreview @DeepIndaba Congratulations, @siminyu_kat!
2022-06-29 21:02:06 @schemeprincess Are those ASICS? Because that's ASIC-looking shoe!(Seriously, those are cool looking!)
2022-06-29 18:48:15 RT @the_transit_guy: Not being able to park directly in front of your destination doesn’t mean your city has a lack of parking.
2022-06-29 11:47:48 @JesseDodge @MSFTResearch @Azure Anyone who uses the "lifetime emissions of five cars" for training a language model based on this paper does not understand it, and that definitely seems to include many researchers.
2022-06-29 11:45:34 @JesseDodge @MSFTResearch @Azure The sentence as written could be interpreted either way, and for the NAS search, I would say they didn't even quantify it, since the emissions estimates were flawed (off by 88X). They did run actual measurements for some of the other numbers in their paper, which confuses things.
2022-06-29 10:13:47 RT @adityakusupati: IntroductingMatryoshka Representations for Adaptive DeploymentTL
2022-06-29 10:08:06 @JesseDodge @MSFTResearch @Azure Is that data sufficient to extend your Figure 1 with GCP data, at least for using A100s which are available in both Azure and GCP?
2022-06-29 10:05:22 @JesseDodge @MSFTResearch @Azure There's pretty decent emissions for GCP regions available at:https://t.co/3wKYn9mdMSAnd maybe in an easier-to-process format in https://t.co/zb4HBaRmsu
2022-06-28 20:58:28 RT @ch402: I'm excited to *finally* be making progress on understanding the first MLP layer in large transformer LMs.I've tried really ha…
2022-06-28 20:19:36 @JesseDodge @MSFTResearch @Azure ... But in a similar way to picking cloud regions appropriately inn order to minimize emissions, it might be worth mentioning that the choice of cloud provider can make a major difference, as can the choice of accelerator hardware.
2022-06-28 20:18:22 @JesseDodge @MSFTResearch @Azure Did you consider evaluating other cloud providers and other ML accelerators? I suspect that GCP would look fairly favorable from a CO2e emissions perspective, as would TPUs compared with A100 GPUs. Seems like that comparison was out of scope for the paper (which is fine)..
2022-06-28 18:57:38 10M Kaggle community members! https://t.co/0qupLOJgkd
2022-06-28 17:51:47 RT @Sifill_LDF: I’m gonna say this before things get out of hand on legal SCOTUS. B/c it is true that many brilliant attys &
2022-06-27 19:14:26 RT @the_transit_guy: [THREAD] Want to see what's possible when you get freeways out of downtown? When Providence relocated its I-195 freewa…
2022-06-27 09:20:54 @Grady_Booch @TheHighNibble The first computer I ever used!
2022-06-26 10:01:11 RT @AndyKimNJ: Yesterday we stood with each other in response to the backsliding of rights. But what about today? What about tomorrow? Wha…
2022-06-25 23:16:01 RT @MoMath1: This month's #MeetAMathematician special guest is @angela_tabiri, an AIMS-Google AI Postdoctoral Fellow at @AIMSGhana! Join An…
2022-06-25 18:59:02 RT @AIMSGhana: Congratulations to our class of 2022! You have made it &
2022-06-25 15:40:53 @taneemishere @TaliaRinger "Party" is the Parti line.
2022-06-25 13:53:48 Cool bit of scientific history in that email!"This is 5 minutes old with zero cross-checks, so please don't share this just yet." https://t.co/2quxYqUqgX
2022-06-25 10:58:48 @Rajputhprashant Can you try exploring these?https://t.co/M1OwvQxEuGhttps://t.co/UjWrmIl78y
2022-06-24 22:50:56 RT @SusanWojcicki: As a CEO I recognize there are a spectrum of opinions on the SCOTUS ruling today. As a woman, it’s a devastating setback…
2022-06-24 22:50:03 RT @ReverendWarnock: I'm outraged by the Supreme Court's decision. As a pro-choice pastor, I’ll never back down from this fight. Women must…
2022-06-24 18:40:40 @TaliaRinger Not After Parti?
2022-06-23 11:02:56 RT @daanvanesch: Happy to share a new data set for improving language diversity in #NLProc! Our paper "Writing System and Speaker Metadata…
2022-06-23 02:06:50 @thehardesh @TheDeshpande Both models produce high quality images but use pretty different approaches (diffusion model vs. autoregressive model). The research teams are working together to combine the approaches.
2022-06-22 23:01:19 @mrdanieldsouza @CVPR @_cagarwal @sarahookr @ml_collective Congrats!
2022-06-22 22:59:43 There's a nice @GoogleAI blog post today by members of the #Imagen and #Parti teams comparing the two approaches. https://t.co/tRgRSo1JMC
2022-06-22 22:56:47 RT @GoogleAI: Check out Parti (https://t.co/mWTkkp3iVt), our second text-to-image model after the release of our recent Imagen work (https:…
2022-06-22 22:56:31 RT @ZoubinGhahrama1: I'm super excited about our new work on photorealistic text-to-image generation! You may have seen #imagen already...…
2022-06-22 22:56:24 RT @lmthang: Excited to share our work on Parti, state-of-the-art text-to-image model that generates photorealistic images with complex com…
2022-06-22 22:54:05 Also interesting to see outputs of different sizes of the model. Better text, fidelity.."A portrait photo of a kangaroo wearing an orange hoodie and blue sunglasses standing on the grass in front of the Sydney Opera House holding a sign on the chest that says Welcome Friends!" https://t.co/PTBuhuibCB
2022-06-22 22:40:16 RT @GoogleAI: Today we present the first demonstration of a provable exponential advantage in #QuantumMachineLearning over classical algori…
2022-06-22 22:36:09 Finally, it's unclear if this is a holy wombat or not..
2022-06-22 22:34:45 Also, did I mention it's Parti time?
2022-06-22 18:31:54 In exciting new areas like text-to-image generation, it's important to explore multiple research approaches. In Parti and Imagen, we have good understandings of these two approaches and hope to combine the best aspects of each in future work.
2022-06-22 18:30:43 Parti is new work from @GoogleAI, and it stands for the Pathways Autoregressive Text-to-Image model. The autoregressive approach used in Parti is different than the diffusion model approach in the Imagen work released a few weeks ago.(cont)
2022-06-22 18:27:27 "A photo of the back of a wombat wearing a backpack and holding a walking stick. It is next to a waterfall and is staring at a distant mountain." #partihttps://t.co/aPI43xa7v1 https://t.co/B5DjicsRU8
2022-06-22 01:25:11 Leggo my Eggo, corporate edition https://t.co/oiulXLF8SF
2022-06-20 20:15:17 RT @RishiBommasani: We just put out a paper on emergent abilities of LMs, led by the vivacious @_jasonwei! https://t.co/verGuHwjJ6@_jasonw…
2022-06-16 03:19:32 Bleh. Painstakingly found Twitter handles and then messed up on @LiamFedus's, which should have been the easiest!
2022-06-16 02:50:16 An analysis of emergent properties of large language models. Joint work w/ @_jasonwei, @YiTayML, @RishiBommasani, @colinraffel, @barret_zoph, @borgeaud_s, @DaniYogatama, @MaartenBosma, @denny_zhou, @metzlerd, @edchi, @tatsu_hashimoto , @OriolVinyalsML, @percyliang, @liam_fedus https://t.co/9ocqPk2cnP
2022-06-15 19:52:28 For seasoned conference-goers, please weigh in to @sydgibs' thread about how to make the most of a first academic in-person CS conference experience. https://t.co/tSOXzqNjNO
2022-06-15 16:14:54 @adjiboussodieng @VertexLab_ Great name! Meaningful to the work you do, but edgy!
2022-06-15 16:12:07 @HilaCodes @wilbowma Probably @TaliaRinger should as well.
2022-06-15 14:44:47 @TaylorLorenz I wonder if one can hire a dentist on retainer. The retainer fees might be lower.
2022-06-15 14:40:08 RT @thomaskipf: So excited to share Object Scene Representation Transformer (OSRT):OSRT learns about complex 3D scenes &
2022-06-15 13:54:29 @sydgibs @PLDI Meet lots of people! Watch the talks but don't feel like you have to attend every last one of them (note presentation styles of speakers you found most effective). Browse posters. Find people w/similar interests, as you'll likely run into them or their work in future. Have fun!
2022-06-15 06:21:02 @HuaizuJiang @Miles_Brundage Very sorry to hear this. His work was quite impactful.
2022-06-13 14:40:21 RT @nathanbenaich: In May 2021, I wrote an @FT oped to catalyse a discussion on rewriting the university spinout playbook.We've since cro…
2022-06-13 04:53:05 RT @sundarpichai: It's exciting to see the spirit of entrepreneurship and innovation happening across Latin America. Proud to support it wi…
2022-06-12 04:38:34 @JohnMcCullough_ Don't forget the ingrates.
2022-06-11 03:44:48 @TaliaRinger Congrats!
2022-06-10 16:24:23 @TaliaRinger I think it was @DynamicWebPaige (?). And I, too, love the bears.
2022-06-10 16:22:25 RT @rikelhood: We present LIMoE, the first large multimodal sparse model! By training routers to find good paths for image &
2022-06-10 02:06:56 BIG-bench is a new set of benchmarks and evaluations of many language models of various scales, lovingly evaluated in exquisite detail in 100 pages by @jaschasd and 441 collaborators at 132 institutions. Great work by all the contributors! https://t.co/oxI1pPwgFZ
2022-06-10 01:57:09 @Ritwik_G @ai_climate Hard to tell for sure if it will work for your use case, but @tanyambirch might be a good person to ask.
2022-06-09 21:51:51 RT @GoogleAI: Read all about LIMoE (the Language Image Mixture of Experts), the first large-scale architecture that processes both images a…
2022-06-09 17:30:29 Dynamic World is a great resource to build tools and applications that depend on evaluating rapidly changing or temporal conditions of the earth.Example below showing before and after for the Caldor fire area in California. https://t.co/B1uDhT8nxD https://t.co/2ZlKYyIsSz
2022-06-09 04:58:08 RT @black_in_ai: Wanna learn priceless skills, expand your network and interact with cutting edge research? Come volunteer with us! We have…
2022-06-08 18:28:51 @DougAndres @StevenTDennis Pallbearers?
2022-06-08 18:27:10 RT @A__Diack: ExploreCSR 2022 is now open for applications."Strong applications will demonstrate how the proposed workshop builds student…
2022-06-08 16:14:17 @vukosi @GoogleAI Congratulations! This is well deserved and great to see!
2022-06-08 04:46:39 @judyfhoffman @GoogleAI @ICatGT @mlatgt Congrats! This is very well deserved!
2022-06-08 04:30:12 @dataandpolitics Google is hiring in Atlanta for a very large number of roles (as of today, 513 unique roles, 321 in engineering).https://t.co/sg4O9vMXu3
2022-06-08 04:16:03 @TaliaRinger Some sage advice: it's not about having time, it's about making time
2022-06-07 04:58:32 RT @black_in_ai: Are you excited to be meeting in person again to discuss AI, Machine Learning and topics related to Black in AI? Come join…
2022-06-07 04:19:05 RT @GoodNewsCorres1: Retiring after 50 years of teaching. On her last day working at the school, this teacher gets a beautiful send off by…
2022-06-07 04:14:59 RT @adjiboussodieng: Halting my Twitter break to make this post:Are you:Of African descent And working in #AI And in your postdoc…
2022-06-07 04:10:19 RT @googlestudents: Did you know that Google offers scholarships to attend select conferences? And this year's scholarship covers travel …
2022-06-06 16:14:34 @hardmaru The one on the right is more appealing.
2022-06-06 01:54:02 @AnimaAnandkumar @Caltech @bjenik Congratulations, @AnimaAnandkumar and @bjenik!
2022-06-05 04:13:25 RT @pushtheneedle: the empire state building took 400 days to design, permit, and construct. from the first announcement of the project to…
2022-06-04 16:57:57 GitHub: https://t.co/ne1wJEYH8wPaper with more details:Vectorized and performance-portable Quicksort, by Mark Blacher, Joachim, Giesen Peter Sanders, and Jan Wassenberghttps://t.co/j7Z14PIIr5
2022-06-04 16:30:58 Feeling out of sorts? Check out this blog post about a high performance vectorized quicksort implementation by @GoogleAI researcher Jan Wassenberg. https://t.co/t6xduixyPG
2022-06-03 05:22:06 RT @philritz1: @pushtheneedle https://t.co/18ld057lxr
2022-06-03 04:32:30 Nice thread by @borisdayma summarizing the trials &
2022-06-03 04:09:11 Great link to a blog post about the many advantages of using JAX! https://t.co/PBc8elJkmV
2022-06-02 19:40:25 @TaliaRinger @asayeed Señor Barista Fellow.
2022-06-02 02:47:38 @TaliaRinger @ChrSzegedy Bears watching(Very different meaning if you add an apostrophe, I guess)
2022-06-01 00:53:43 And, I'll highlight tweet 18/18 in @iseeaswell's thread, which has a nice map of where these languages are spoken, as well as a pointer to a playlist of music of songs in these languages.https://t.co/Zo1bLhAy3q
2022-06-01 00:50:56 In addition to @iseeaswell's Twitter thread, there's a nice blog post by @iseeaswell and @ankurbpn about the work:https://t.co/vG9Rf4i7ft
2022-06-01 00:42:06 Some more detailed context on the technical work @GoogleResearch that enabled this, which includes training large translation models on 1000 languages simultaneously, &
2022-06-01 00:37:03 Very nice write-up focusing on the recent addition of Quechua as one of 24 new languages added to Google Translate and the impact it has on that community.The 24 new languages added have 300M native speakers."allinllachu" to all. https://t.co/tRkwmD93ev
2022-05-31 18:59:33 RT @IMordatch: How can we effectively train generalist multi-environment agents? We trained a single Decision Transformer model to play man…
2022-05-30 16:47:55 RT @Bob_Wachter: SF folks: it's now a big-time surge. No longer just cases (~500/d reported, so true # >
2022-05-30 05:52:14 RT @sirbayes: @frankdonaldwood @sama @demishassabis @ylecun You asked leaders at OpenAI, Meta, Deepmind who will scale this kind of work -…
2022-05-29 22:02:06 TIL. https://t.co/6wn9NP4rJ0
2022-05-29 20:25:29 @RexChapman Force of habit..
2022-05-29 19:49:25 RT @dpkingma: It was 16 years ago, in 2006, that @geoffreyhinton et al released their demo of deep belief nets. Undergrad me was highly imp…
2022-05-29 19:48:29 @DynamicWebPaige It's hopping!
2022-05-29 17:54:35 @socialistdogmom I relate to this.
2022-05-29 17:49:49 @DynamicWebPaige The whole sequence will be back propagation?
2022-05-29 17:21:42 @LeonDerczynski Given that we've published the paper, open sourced the code and are going to open source the trained checkpoints soon, I'm not sure why you feel this "can't be replicated". I'm happy to hear suggestions about how to make it more replicatable, though.https://t.co/a5J8fcQx2Z
2022-05-29 05:34:27 @YoujunHu @LeonDerczynski Including data on the amount of compute is a requirement of some conference submission checklists. For example:"(d) Did you include the amount of compute and the type of resources used (e.g., type of GPUs, internal cluster, or cloud provider)?..."https://t.co/rTlL6eetdd
2022-05-28 23:04:57 @_joaogui1 Perhaps something done by Carver Mead and collaborators?https://t.co/DzUSL6VUM2
2022-05-28 22:49:45 @LeonDerczynski @agesmundo We are also releasing the trained checkpoints for the experiment in the paper (waiting for 2 people to flip approval bits).
2022-05-28 22:49:12 @LeonDerczynski @agesmundo The code has been open-sourced for those who want to re-run the experiments (acknowledging that it will require a fair amount of compute to do so).https://t.co/e83bMbHzS7
2022-05-28 22:32:15 @bakztfuture @LeonDerczynski Publications produced by external researchers using compute resources provided by the TPU Research Cloud program:https://t.co/R4msUmWwjl
2022-05-28 22:31:23 @bakztfuture @LeonDerczynski We already do devote considerable resources to this through our TPU Research Cloud program:https://t.co/0t3ylcx622That has 1,000 TPU devices (4-8 chips each) with a mix of TPUv2, TPUv3, and TPUv4 chips (I believe it's about ~0.5 exaflops of total peak compute).
2022-05-28 22:26:42 I responded to original tweet, but since this is making the rounds, goal of the work was not to train a high quality cifar10 model, but rather to explore learning on stream of 69 tasks w/o catastrophic forgetting (w/sparsity &
2022-05-28 22:15:48 RT @JeffDean: @LeonDerczynski The goal of this work was not to train a cifar10 model, but rather to explore multi-task learning on a stream…
2022-05-28 22:12:23 @LeonDerczynski Also, Andrea is on Twitter.cc: @agesmundo
2022-05-28 22:10:23 @LeonDerczynski Bleh. s/catastrapohic/catastrophic/
2022-05-28 22:08:33 @LeonDerczynski Example video of the system:https://t.co/C325cNcRFhWe really should set these to music, but for now, I recommend watching on 2X speed :)
2022-05-28 22:07:05 @LeonDerczynski The goal of this work was not to train a cifar10 model, but rather to explore multi-task learning on a stream of 69 tasks while avoiding catastrapohic forgetting, and the cost numbers in the reddit post are also off considerably. I responded on reddit:https://t.co/Oq2GRCpmnP https://t.co/UHDDNUDI0t
2022-05-28 21:59:10 @zek3r The goal of this work was not to train a cifar10 model, and the cost numbers are significantly off. I responded on Reddit.Links to the videos mentioned below:https://t.co/C325cNcRFhhttps://t.co/ao1G7dyf88Reddit link:https://t.co/Oq2GRCpmnP https://t.co/k410yDisoi
2022-05-28 04:21:25 RT @QueerinAI: Queer in AI is coming to NeurIPS 2022! We would be excited to have you onboard as volunteers and join us in organizing NeurI…
2022-05-28 04:14:08 @savvyRL "Prompt delivery driver"
2022-05-27 05:23:58 @berkeley_ai @Stanford @DorsaSadigh @chelseabfinn @ieeeras Congratulations, @DorsaSadigh and @chelseabfinn! Very well deserved!
2022-05-26 19:57:26 RT @A__Diack: We're excited about this partnership and the potential of joint research collaboration with AIMS.Yossi Matias, Google's VP…
2022-05-26 19:57:20 RT @AIMS_Next: We're thrilled to announce that we signed a Masters Sponsored Research Agreement (MSRA) with @Google Research through @AIMSG…
2022-05-26 19:39:51 Some really neat work by @Yuhu_ai_ and others showing that large language models can turn informal natural language mathematical statements into formal specifications. https://t.co/7mvItLUrqF
2022-05-26 05:36:32 @mo_norouzi @GarethLCollins This prompt has served its porpoise.
2022-05-26 05:30:25 @Chitwan_Saharia @Maajora_ @mo_norouzi We've been ghosted.
2022-05-26 05:28:23 @Chitwan_Saharia @DCasBol @mo_norouzi Czech and mate, honey badger!
2022-05-26 03:52:53 If you want to see many examples of images generated by the #imagen model based on user requests, then this is the thread for you. https://t.co/GTojbtmbiq
2022-05-26 00:21:28 @dave_andersen I'm so sorry to hear this, Dave.
2022-05-25 14:19:50 RT @irishrygirl: I can buy 10,000 rounds of ammo and there is no background check, but if I buy 1 box of Sudafed my info goes into a nation…
2022-05-25 13:33:48 RT @A__Diack: Google Cloud Computing Foundations Program is now available to faculty in Cameroon, Ethiopia, Rwanda, Zimbabwe and Ivory Coas…
2022-05-25 13:33:22 RT @A__Diack: Google Resources for African Researchers and Academics 1/10PhD Fellowship ProgramThe Google PhD Fellowship supports outs…
2022-05-25 06:30:32 RT @LosPollosTV: Wtf am I even looking at? https://t.co/XNSQmi4MjZ
2022-05-25 05:37:08 RT @tabstrr1: Unfortunately my beautiful Nevaeh was one of the many victims from todays tragedy. Thank you for the support and help. Rest i…
2022-05-25 05:16:56 RT @googlecanada: Today, the @Google Research Brain Team in Toronto introduces Imagen, a photorealistic text-to-image model! Here:"Androi…
2022-05-25 05:14:19 @tabstrr1 @DocJeffD I am so sorry for your loss, and for the world's loss. The senselessness of gun violence in this country affects so many and is a needless tragedy: we can and should reform our laws to prevent such events. It is in our collective power.
2022-05-25 04:57:46 RT @NotoriousOHM: A furious, emotional and fed up Steve Kerr pleaded with senators to do something about the mass shootings. https://t.co/f…
2022-05-25 04:52:08 RT @joncoopertweets: Who else managed to live their entire lives without needing to own an assault rifle?
2022-05-25 04:37:41 @ageshah @GoogleAI Hard to say. I do think some of it is in the prompt, because we're asking the model to capture a candid moment of two meerkats, which is slightly whimsical. It's also more interesting to use prompts that are somewhat unrealistic (panda bear chefs, mugs floating in chocolate, ...)
2022-05-24 23:45:37 @TomLikesRobots @GoogleAI Figure 2 of the paper has non-cherry-picked examples for DrawBench prompts:https://t.co/Gw3el4F6rc
2022-05-24 23:39:06 RT @Chitwan_Saharia: We are thrilled to announce Imagen, a text-to-image model with unprecedented photorealism and deep language understand…
2022-05-24 19:10:55 Also, those are not mere cats in the first image.
2022-05-24 19:09:29 The website at https://t.co/JEgyNrcJjl has more examples and some interactive demos where you can play with selecting some phrases to form a (constrained) sentence.It also has some good discussion around the Limitations and Societal Impact of these sorts of models. https://t.co/eDc1QJbxDl
2022-05-24 19:05:07 https://t.co/Tig9fK0fm7
2022-05-24 19:04:49 https://t.co/xM7235NjoR
2022-05-24 19:04:31 Be sure to check out the paper at:https://t.co/Gw3el4F6rcThis is great work by @Chitwan_Saharia, @wchan212, @srbhsxn, Lala Li, @jaywhang_, @cephaloponderer, @coolboi95, Burcu Karagol Ayan, Sara Mahdavi, @iraphas13, @TimSalimans, @hojonathanho, @fleet_dj, and @mo_norouzi. https://t.co/M5P08LFSvI
2022-05-24 18:51:21 New @GoogleAI work:Input: "Two meerkats sitting next to each other on top of a mountain and looking at the beautiful landscape. There is a mountain, a river lake, and fields of yellow flowers. There are hot air balloons in the sky."#imagen https://t.co/JEgyNrcJjlOutput: https://t.co/uj4urjnZPF https://t.co/I1zx8ZARBl
2022-05-24 18:43:42 Be sure to check out the paper at:https://t.co/Gw3el4nvzEThis is great work by @Chitwan_Saharia, @wchan212, @srbhsxn, Lala Li, @jaywhang_, @cephaloponderer, @coolboi95, Burcu Karagol Ayan, Sara Mahdavi, @iraphas13, @TimSalimans, @hojonathanho, @fleet_dj, and @mo_norouzi.
2022-05-24 05:37:27 @adityagrover_ Alternatively, it's a real image of a panda chef preparing bamboo-infused croissant.
2022-05-20 08:11:00 CAFIAC FIX
2022-10-28 23:11:20 @kelseyhightower Congratulations, Kelsey! You've more than Distinguished yourself over the years!
2022-10-28 20:29:39 I'm excited to see our @GoogleAI researchers share how Google’s AI technologies can help people, particularly to stay safe, better connect, and be creatively inspired.Join us on Nov 2nd to watch AI@ live! https://t.co/ug7ac8GDTk
2022-10-28 03:12:36 @balajiworld No embarrassment! Different opportunities present themselves over time, and the important thing is to be accepting of others and to be curious when you do get the chance to travel.
2022-10-28 03:07:57 @minhtrann @dave_andersen First and second grade, near Little Rock. We lived on five acres that was a former chinchilla ranch (parents remodeled main building into a house), &
2022-10-27 11:40:06 @lamechthinkbig @dave_andersen I wish there was (I'm not involved in the project, but there's a GitHub link for it)
2022-10-27 02:08:55 @jbrownkramer @dave_andersen I guess that's a goal in the search of the state space.
2022-10-27 02:04:15 @jbrownkramer @dave_andersen Sorry! I actually want to visit all the states I haven't been to!
2022-10-27 02:01:52 @lk_bell @dave_andersen I did it as "stayed == spent at least one night there"
2022-10-27 02:01:08 https://t.co/OFvpBZQUgl
2022-10-27 01:57:59 (Feel free to reply with yours! There's also an EU version)
2022-10-27 01:50:42 RT @GoogleAI: On Nov 2 at 9am EST, we're excited to host our first AI@ event: a showcase of our latest and greatest work in AI. Tune in liv…
2022-10-27 01:45:42 This was kind of fun (although like @dave_andersen, I suspect I'm forgetting a few road trips or short stops here and there).https://t.co/JsFkJzWEUZ https://t.co/JKnAPP4zDl
2022-10-26 14:11:08 @moyix @danrkports You can use net, but not use nets.
2022-10-25 21:46:52 RT @GoogleAI: Learn how Open Images V7 expands the Open Images dataset with new point-level label annotations, which provide localization i…
2022-10-25 03:27:43 @wowinteresting8 Please power cycle your phone
2022-10-23 21:34:41 RT @quocleix: New open-source language model from Google AI: Flan-T5 Flan-T5 is instruction-finetuned on 1,800+ language tasks, leading…
2022-10-23 17:25:01 If you're at #ECCV2022 (virtually or physically), @GoogleAI researchers that are also attending would be happy to chat with you (about their work, your work, career advice, and many other topics!). See below for list of 60+ accepted Google-affiliated papers + workshops/tutorials. https://t.co/3FRcQsTadv
2022-10-22 21:44:11 A recent blog post describing some of the @TAE/@GoogleAI collaboration.https://t.co/piVYFpQgPk
2022-10-22 20:29:08 Nice article about the effort to develop fusion energy, focused on @TAE. Amazing for the world if we can do it! @GoogleAI has a many-year collaboration w/TAE to use machine learning to aid in creating stable fusion reactions.https://t.co/JDyFXORKT7
2022-10-22 20:18:40 RT @sundarpichai: Work from our @GoogleAI team - training robots to play table tennis to better understand how they learn in dynamic + high…
2022-10-22 12:41:33 @RisingSayak @ksoonson I'm glad you liked the work!
2022-10-21 22:13:13 @dave_andersen I'm willing to have the "In my day, when I was your age, my acoustic coupler topped out at 110 baud" discussion, if helpful.
2022-10-21 21:31:32 RT @BrentToderian: Decades ago, many of us urbanists had this image in our slide deck to illustrate how simple exercise has been designed o…
2022-10-21 02:46:28 @bonadossou @mcgillu @McGill_Eng @McGill_CIM Congratulations!
2022-10-20 20:11:40 RT @rikelhood: The recordings from the Google Workshop on Sparsity and Adaptive Computation are now available at:Day 1: https://t.co/8yUH…
2022-10-20 13:00:42 RT @kelvin_guu: New from Google Research! Language models perform amazing feats, but often still "hallucinate" unsupported content. Our mod…
2022-10-19 18:43:00 RT @JeffDean: @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. Fo…
2022-10-19 17:47:57 @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. For some reason it wasn't showing up on this list of articles by the author, but I found it at this URL:https://t.co/pkAQ0IRmBj
2022-10-19 11:30:13 RT @GoogleAI: Applications are open for CSRMP, a mentorship program that matches students from historically marginalized groups in computin…
2022-10-19 11:29:49 RT @GoogleAI: Learn about AudioLM, an audio generation framework that demonstrates long-term consistency (e.g., syntax in speech &
2022-10-19 11:29:40 RT @GoogleAI: Image captioning datasets can be limited for languages other than English. So, today we are releasing the Crossmodal 3600 dat…
2022-10-29 20:03:42 RT @YouAreLobbyLud: This seems amazing. Belgium now requires, BY LAW:1) public indoor spaces must display CO2 level2) risk analysis and…
2022-10-29 19:58:32 @moultano I am here for aleph this..
2022-10-29 19:53:26 @Anon__08 @jsomers @NewYorker Nope..
2022-10-28 23:11:20 @kelseyhightower Congratulations, Kelsey! You've more than Distinguished yourself over the years!
2022-10-28 20:29:39 I'm excited to see our @GoogleAI researchers share how Google’s AI technologies can help people, particularly to stay safe, better connect, and be creatively inspired.Join us on Nov 2nd to watch AI@ live! https://t.co/ug7ac8GDTk
2022-10-28 03:12:36 @balajiworld No embarrassment! Different opportunities present themselves over time, and the important thing is to be accepting of others and to be curious when you do get the chance to travel.
2022-10-28 03:07:57 @minhtrann @dave_andersen First and second grade, near Little Rock. We lived on five acres that was a former chinchilla ranch (parents remodeled main building into a house), &
2022-10-27 11:40:06 @lamechthinkbig @dave_andersen I wish there was (I'm not involved in the project, but there's a GitHub link for it)
2022-10-27 02:08:55 @jbrownkramer @dave_andersen I guess that's a goal in the search of the state space.
2022-10-27 02:04:15 @jbrownkramer @dave_andersen Sorry! I actually want to visit all the states I haven't been to!
2022-10-27 02:01:52 @lk_bell @dave_andersen I did it as "stayed == spent at least one night there"
2022-10-27 02:01:08 https://t.co/OFvpBZQUgl
2022-10-27 01:57:59 (Feel free to reply with yours! There's also an EU version)
2022-10-27 01:50:42 RT @GoogleAI: On Nov 2 at 9am EST, we're excited to host our first AI@ event: a showcase of our latest and greatest work in AI. Tune in liv…
2022-10-27 01:45:42 This was kind of fun (although like @dave_andersen, I suspect I'm forgetting a few road trips or short stops here and there).https://t.co/JsFkJzWEUZ https://t.co/JKnAPP4zDl
2022-10-26 14:11:08 @moyix @danrkports You can use net, but not use nets.
2022-10-25 21:46:52 RT @GoogleAI: Learn how Open Images V7 expands the Open Images dataset with new point-level label annotations, which provide localization i…
2022-10-25 03:27:43 @wowinteresting8 Please power cycle your phone
2022-10-23 21:34:41 RT @quocleix: New open-source language model from Google AI: Flan-T5 Flan-T5 is instruction-finetuned on 1,800+ language tasks, leading…
2022-10-23 17:25:01 If you're at #ECCV2022 (virtually or physically), @GoogleAI researchers that are also attending would be happy to chat with you (about their work, your work, career advice, and many other topics!). See below for list of 60+ accepted Google-affiliated papers + workshops/tutorials. https://t.co/3FRcQsTadv
2022-10-22 21:44:11 A recent blog post describing some of the @TAE/@GoogleAI collaboration.https://t.co/piVYFpQgPk
2022-10-22 20:29:08 Nice article about the effort to develop fusion energy, focused on @TAE. Amazing for the world if we can do it! @GoogleAI has a many-year collaboration w/TAE to use machine learning to aid in creating stable fusion reactions.https://t.co/JDyFXORKT7
2022-10-22 20:18:40 RT @sundarpichai: Work from our @GoogleAI team - training robots to play table tennis to better understand how they learn in dynamic + high…
2022-10-22 12:41:33 @RisingSayak @ksoonson I'm glad you liked the work!
2022-10-21 22:13:13 @dave_andersen I'm willing to have the "In my day, when I was your age, my acoustic coupler topped out at 110 baud" discussion, if helpful.
2022-10-21 21:31:32 RT @BrentToderian: Decades ago, many of us urbanists had this image in our slide deck to illustrate how simple exercise has been designed o…
2022-10-21 02:46:28 @bonadossou @mcgillu @McGill_Eng @McGill_CIM Congratulations!
2022-10-20 20:11:40 RT @rikelhood: The recordings from the Google Workshop on Sparsity and Adaptive Computation are now available at:Day 1: https://t.co/8yUH…
2022-10-20 13:00:42 RT @kelvin_guu: New from Google Research! Language models perform amazing feats, but often still "hallucinate" unsupported content. Our mod…
2022-10-19 18:43:00 RT @JeffDean: @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. Fo…
2022-10-19 17:47:57 @Ket_Cherie @lebovitzben @ylecun @dmonett @Grady_Booch @marklemley Great pointer to an interesting article I hadn't seen. For some reason it wasn't showing up on this list of articles by the author, but I found it at this URL:https://t.co/pkAQ0IRmBj
2022-10-19 11:30:13 RT @GoogleAI: Applications are open for CSRMP, a mentorship program that matches students from historically marginalized groups in computin…
2022-10-19 11:29:49 RT @GoogleAI: Learn about AudioLM, an audio generation framework that demonstrates long-term consistency (e.g., syntax in speech &
2022-10-19 11:29:40 RT @GoogleAI: Image captioning datasets can be limited for languages other than English. So, today we are releasing the Crossmodal 3600 dat…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-22 05:56:08 @docmilanfar The police car really got plowed into by the, uh, plow.
2022-11-21 18:40:23 RT @GoogleAI: Applications for the first-ever Google PhD Fellowships for students in Latin America open today, along with applications to s…
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-22 05:56:08 @docmilanfar The police car really got plowed into by the, uh, plow.
2022-11-21 18:40:23 RT @GoogleAI: Applications for the first-ever Google PhD Fellowships for students in Latin America open today, along with applications to s…
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-23 23:08:35 RT @DH_PlantVillage: #ThanksGiving is about harvests. These women in Tanzania have been barely harvesting cassava due to viral diseases. I…
2022-11-22 05:56:08 @docmilanfar The police car really got plowed into by the, uh, plow.
2022-11-21 18:40:23 RT @GoogleAI: Applications for the first-ever Google PhD Fellowships for students in Latin America open today, along with applications to s…
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-11-29 02:32:24 @FlavianHautbois Pair programming can definitely be used for training when one person has a lot of experience or expertise relative to the other person. I haven't generally used it in this way, but I've seen others do so. I often use code review comments to pass along different tips and tricks.
2022-11-27 00:27:42 @claucece @pati_gallardo @johnregehr @uwcse Sorry, I missed your question when you first sent it. This list seems good and covers a bunch of interesting subtopics: https://t.co/rHM0ghIzhe
2022-11-26 03:32:38 RT @thorpej: Engaged in some #bicycle shenanigans today. https://t.co/9777FQpcpD
2022-11-26 03:27:13 @dataandpolitics I guess it's not a food desert.
2022-11-25 16:44:41 There's a sucker born every minute.. https://t.co/IQUPcv0Wgv
2022-11-23 23:08:35 RT @DH_PlantVillage: #ThanksGiving is about harvests. These women in Tanzania have been barely harvesting cassava due to viral diseases. I…
2022-11-22 05:56:08 @docmilanfar The police car really got plowed into by the, uh, plow.
2022-11-21 18:40:23 RT @GoogleAI: Applications for the first-ever Google PhD Fellowships for students in Latin America open today, along with applications to s…
2022-11-18 06:38:45 Proud to have personally backed @Astranis and @Gedmark. They're on an exciting path working to bring satellite internet to billions of people! I really like that they move faster and think about engineering differently than most traditional satellite makers. https://t.co/HFoIcPRCSV
2022-11-18 06:19:58 RT @oh_that_hat: “LLMs can’t even do addition” We show that they CAN add! To teach algos to LLMs, the trick is to describe the algo in e…
2022-11-18 06:12:57 Very sad news that Fred Brooks has passed away today. I was fortunate to meet him at @HLForum in 2018. He made monumental contributions to the practice of software engineering. His book "The Mythical Man Month: Essays on Software Engineering" is full of timeless observations. https://t.co/WIiLqGnbiG
2022-11-18 05:26:57 Great progress on this problem, @jmgilmer! https://t.co/fW1mLExgkh
2022-11-18 02:18:26 RT @Imposter_Edits: There was a baby chimp born at The Sedgwick County Zoo, the baby had to be put on oxygen for a few days. This is a clip…
2022-11-17 05:09:12 RT @BrentToderian: This is VERY interesting — imagine if every decision-maker in YOUR city, elected or appointed, was required to experienc…
2022-11-16 14:42:10 @AllaYeshwanth Yes: "Undergrad / MS / PhD student in India? Apply before Nov 28."
2022-11-16 03:48:26 Looking forwarding to visiting Bangalore in January! https://t.co/uiJDIf0dKF
2022-11-15 06:38:06 RT @ilove_aviation: Never underestimate a 747! https://t.co/llKSPaGzf4
2022-12-09 06:30:41 @_ejgonz_ @Google @twi_mar Welcome to Google!