Gerard Ben Arous
7 papers · 2015–2024 · 6 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓+3 more ↓ Show less ↑
π£ Hot Topic Early Bird π Academic Marathon Β· Sprint π Interdisciplinary Bridge Β· Bronze πΊοΈ Taxonomy Completionist (14) π Conference Polyglot Β· Silver
π
Keyword Champion
(2)
π
Keyword Trendsetter Combo
(3)
π§
Keyword Pioneer
Top co-authors
Research topics
Machine Learning
(7)
Optimization & Theory Β· Machine Learning
(7)
Mathematics & Optimization
(4)
Deep Learning
(2)
Mathematics
(2)
Probability
(2)
Artificial Intelligence
(1)
Architectures
(1)
Bayesian & Probabilistic Β· Artificial Intelligence
(1)
Bayesian & Probabilistic Β· Machine Learning
(1)
Optimization
(1)
Optimization & Theory Β· Deep Learning
(1)
Keywords
stochastic gradient descent
(4)
neural network
(2)
loss surface
(2)
spin-glass model
(2)
parameter estimation
(1)
high-dimensional inference
(1)
neural network training
(1)
sparse principal component analysis
(1)
non-convex optimization
(1)
high-dimensional analysis
(1)
free energy
(1)
hessian matrix
(1)
phase retrieval
(1)
eigenvalue analysis
(1)
gradient flow
(1)
markov chain monte carlo
(1)
generalized linear model
(1)
stochastic differential equation
(1)
training dynamics
(1)
random matrix theory
(1)
Papers
High-dimensional SGD aligns with emerging outlier eigenspaces
ICLR 2024
High-dimensional limit theorems for SGD: Effective dynamics and critical scaling
NeurIPS 2022
Online stochastic gradient descent on non-convex losses from high-dimensional inference
JMLR 2021
Free Energy Wells and Overlap Gap Property in Sparse PCA
COLT 2020
Comparing Dynamics: Deep Neural Networks versus Glassy Systems
ICML 2018
Open Problem: The landscape of the loss surfaces of multilayer networks
COLT 2015
The Loss Surfaces of Multilayer Networks
AISTATS 2015