Luca Arnaboldi
3 papers · 2023–2024 · 2 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 ↓
🌍
Conference Polyglot · Bronze
🌉
Interdisciplinary Bridge · Bronze
Top co-authors
Research topics
Keywords
information exponent
(2)
gradient descent
(1)
neural network dynamics
(1)
gradient flow
(1)
mean-field limit
(1)
two-layer network
(1)
batch size
(1)
high-dimensional limit
(1)
high-dimensional regime
(1)
sufficient statistics
(1)
two-layer neural network
(1)
mean-field regime
(1)
dynamical mean field theory
(1)
batch reuse
(1)
leap exponent
(1)
multi-pass
(1)
correlation loss sgd
(1)
iteration time
(1)
stochastic gradient descent
(1)
one-pass stochastic gradient descent
(1)
Papers
Online Learning and Information Exponents: The Importance of Batch size & Time/Complexity Tradeoffs
ICML 2024
The Benefits of Reusing Batches for Gradient Descent in Two-Layer Networks: Breaking the Curse of Information and Leap Exponents
ICML 2024
From high-dimensional & mean-field dynamics to dimensionless ODEs: A unifying approach to SGD in two-layers networks
COLT 2023