QLS Seminar - Wednesday, 5 November at 11h00 "Statistical physics of deep learning: Optimal learning of a multi-layer perceptron near interpolation" by Jean Barbier

[Quantitative Life Sciences]

Dear All,

On Wednesday, 5 November at 11:00 CET,  Jean Barbier(QLS and Mathematics 
Sections, ICTP) will give a seminar titled:

*"Statistical physics of deep learning: Optimal learning of a 
multi-layer perceptron near interpolation"

*Abstract:
For three decades, statistical physics has framed neural-network 
analysis, but its reach to expressive, feature-learning deep models was 
unclear. We answer yes by studying supervised learning in fully 
connected multi-layer nets whose hidden layers scale with input 
dimension—favoring feature learning over ultra-wide kernels while 
remaining more expressive than narrow or fixed-weight models—in the 
challenging interpolation regime where parameters and data are 
comparable. Using a matched teacher–student setup, we characterize 
fundamental performance limits and the sufficient statistics learned as 
data grows. The analysis uncovers rich phenomenology with multiple 
learning transitions: with enough data, optimal performance requires 
“specialization” of the student to the target, yet practical training 
can be trapped in sub-optimal solutions. Specialization is 
inhomogeneous—spreading from shallow to deep layers and unevenly across 
neurons—and deeper targets are intrinsically harder. Though derived in a 
Bayesian-optimal setting, the insights on nonlinearity, depth, and 
finite (proportional) width likely generalize.


The seminar will take place in the Common area, Old SISSA building, 
second floor - Via Beirut, 2

Indico: https://indico.ictp.it/event/11217/

You are all most welcome to attend!

Best regards,

Erica

Erica Sarnataro
Group Secretary
Quantitative Life Sciences
The Abdus Salam International Centre for Theoretical Physics (ICTP)
Trieste,  Italy
Tel. +39-040-22404623 (NEW PHONE NUMBER)
www.ictp.it/research/qls.aspx
e-mail:qls at ictp.it