SMS scnews item created by Hannah Bryant at Mon 20 Mar 2023 1200
Type: Seminar
Modified: Mon 20 Mar 2023 1324; Mon 20 Mar 2023 1325
Distribution: World
Expiry: 25 Mar 2023
Calendar1: 23 Mar 2023 1300-1400
CalLoc1: Quad S249 & Online
CalTitle1: SMRI Seminar: Yang ’The unreasonable effectiveness of mathematics in large scale deep learning’
Auth: (hbry8683) in SMS-SAML

SMRI Seminar: Yang -- The unreasonable effectiveness of mathematics in large scale deep learning

SMRI Seminar 
’The unreasonable effectiveness of mathematics in large scale deep learning’ 
Greg Yang (Microsoft Research in Redmond, Washington) 

Thursday 23rd March 1:00-2:00pm 
Quad 249 & Online 

Abstract: Recently, the theory of infinite-width neural networks led to the first
technology, muTransfer, for tuning enormous neural networks that are too expensive to
train more than once.  For example, this allowed us to tune the 6.7 billion parameter
version of GPT-3 using only 7% of its pretraining compute budget, and with some
asterisks, we get a performance comparable to the original GPT-3 model with twice the
parameter count.  In this talk, I will explain the core insight behind this theory.  In
fact, this is an instance of what I call the *Optimal Scaling Thesis*, which connects
infinite-size limits for general notions of "size" to the optimal design of large
models in practice, illustrating a way for theory to reliably guide the future of AI.
I’ll end with several concrete key mathematical research questions whose resolutions
will have incredible impact on how practitioners scale up their NNs.  

Biography: Greg Yang is a researcher at Microsoft Research in Redmond, Washington.  He
joined MSR after he obtained Bachelor’s in Mathematics and Master’s degrees in Computer
Science from Harvard University, respectively advised by ST Yau and Alexander Rush.  He
won the Hoopes prize at Harvard for best undergraduate thesis as well as Honorable
Mention for the AMS-MAA-SIAM Morgan Prize, the highest honor in the world for an
undergraduate in mathematics.  He gave an invited talk at the International Congress of
Chinese Mathematicians 2019.  


Upcoming SMRI events can be found here: 

SMRI YouTube Channel:

ball Calendar (ICS file) download, for import into your favourite calendar application
ball UNCLUTTER for printing
ball AUTHENTICATE to mark the scnews item as read
School members may try to .