Robocrunch        AI

Quanta Magazine    @QuantaMagazine   ·   9/14/2021
An abstract space is n-dimensional if there are n degrees of freedom within it, or if it requires n coordinates to describe the location of a point. This sounds simple, but the idea of dimension can still be devilishly complex.
 Reply      Retweet   20      Like     46    

  Similar Tweets  

Joscha Bach    @Plinz   ·   8/21/2021
Platforms and governments have to fight disinformation. But in a liberal society, it is up to the individual to build and decide about reputation. There is no freedom to think without the freedom to consider every argument, and freedom of speech requires freedom to listen.
 Reply      Retweet   16      Like     156    

Matt Fontaine    @tehqin17   ·   9/14/2021
Amazed at how such complex levels emerge from such small and simple NCAs trained via CMA-ME, where the constraints and variation are defined from very simple functions. No target data required!
 Reply      Retweet   1      Like     4    

Yann LeCun    @ylecun   ·   6/29/2021
This is another piece that basically says "deep learning is not as impressive as you think because it's mere interpolation resulting from glorified curve fitting". But in high dimension, there is no such thing as interpolation. In high dimension, everything is extrapolation.
 Reply      Retweet   159      Like     1079    

David Pfau    @pfau   ·   9/11/2021
And real progress in AI will probably not come until we have the equivalent of complex analysis for intelligence.
 Reply      Retweet        Like     13    

Samarth Sinha    @_sam_sinha_   ·   9/14/2021
Our paper on “Surprisingly Simple Self-Supervision for Offline RL” got accepted at CoRL 2021! Find out how simple data augmentations from states can help Q-learning algorithms on a variety of robotics tasks! J/w: @AjayMandlekar @animesh_garg
 Reply      Retweet   1      Like     3    

Alex Kontorovich    @AlexKontorovich   ·   7/8/2021
7/ Nash's coordinates exist abstractly, but it's not necessarily (we'll see what String Theory etc eventually amounts to... if anything...) the case that there really exist "extra" dimensions for describing the manifold of our Universe. That's my two cents...
 Reply      Retweet   1      Like     4    

MathType    @MathType   ·   8/16/2021
The #Lorentz Transformations are the analogous of the Galilean Transformations for special #relativity, relating the position and time coordinates in two inertial reference frames. #MathType
 Reply      Retweet   42      Like     183    

Posted by:

VentureBeat    @VentureBeat   ·   9/7/2021
Large language models aren't always more complex by @Kyle_L_Wiggers
 Reply      Retweet   2      Like     2    

Dan Roy    @roydanroy   ·   8/17/2021
Not sure about the "pivoting to position themselves as ...". Sounds like startup language, not academics.
 Reply      Retweet   1      Like     51    

Posted by:
Dan Roy

Mark Suster    @msuster   ·   9/14/2021
14/ If you look at our industry, it's now awash with capital that requires every firm to get very focused on where its edge comes from. If you're not a bulge-bracket VC, choose your positioning wisely (stage, sector, geography, founder access) ... way too many firms are generic
 Reply      Retweet   1      Like     5    

MichiganAI    @michigan_AI   ·   9/14/2021
Call for POSTERS! Do you have work in #AI that you would like to share virtually? Participate with a poster in the Michigan AI Symposium. Submit the title/abstract by Oct. 15. Registration also required (free).
 Reply      Retweet   2      Like     2    

Xiang Ren    @xiangrenNLP   ·   6/11/2021
📢 We'd like to hear position & perspectives talks that critically examine common practices & trends on deep learning. Less than 3 weeks to submit an extended abstract for your talk at the KDD 2021 Deep Learning Day (DDL: July 1st). 👇call & submission
 Reply      Retweet   9      Like     21    

Posted by:
Xiang Ren

Jonathan Frankle    @jefrankle   ·   6/9/2021
REMINDER: Sparsity in Neural Networks workshop deadline is in less than a week! You just need to submit an abstract and any supporting materials you like (blog post/paper/poster/slides). Full information here:
 Reply      Retweet   6      Like     33    

Giorgio Patrini    @GiorgioPatrini   ·   5/28/2021
The paper abstract: "We use transfer learning" What they actually mean: "pretrained=True" It didn't use to be like that. It's amazing how Transfer is now a machine learning commodity (in some domains)
 Reply      Retweet   2      Like     7    

  Relevant People  

Quanta Magazine
Big ideas in science and math. Because you want to know more. Launched by @SimonsFdn. Get our weekly newsletter:
Quanta Magazine 56.7

Matt Fontaine
roboticist in training @ USC, Algorithms Live! host, ICPC judge, USACO coach
Matt Fontaine 7.63

Joscha Bach
FOLLOWS YOU. Artificial Intelligence, Cognitive Architectures, Computation. The goal is integrity, not conformity.
Joscha Bach 45.3

Danijar Hafner
Building unsupervised intelligent machines embodied in complex environments, via world models & infomax. Researcher @GoogleBrain and PhD student @UofT.
Danijar Hafner 32.4

Graham Neubig
Associate professor at CMU, studying natural language processing, machine learning, etc. Japanese account is @neubig.
Graham Neubig 40.1

Type and handwrite mathematical notation. Easily include quality math equations in your documents and digital content. Part of @wiris_science.
MathType 34.1

Samarth Sinha
Researching ML @FacebookAI & @VectorInst Prev: Undergrad @UofT, Visitor @Mila_Quebec @Berkeley_AI
Samarth Sinha 21.4