Benoit Dherin
6 papers · 2021–2024 · 3 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
🗺️
Taxonomy Completionist
(18)
Venues
Top co-authors
Research topics
Machine Learning
(6)
Optimization & Theory · Machine Learning
(6)
Artificial Intelligence
(3)
Deep Learning
(3)
Optimization & Theory · Deep Learning
(3)
Core AI
(2)
Natural Language Processing
(1)
Learning Paradigms
(1)
Learning Types · Machine Learning
(1)
Learning Types · Deep Learning
(1)
Resources & Methods
(1)
Keywords
implicit regularization
(3)
gradient descent
(2)
backward error analysis
(2)
geometric complexity
(2)
learning rate
(1)
neural collapse
(1)
gradient flow
(1)
foundation model
(1)
generative adversarial network
(1)
training dynamics
(1)
parameter initialization
(1)
flat minima
(1)
few-shot setting
(1)
generalization performance
(1)
two-player game
(1)
dirichlet energy
(1)
mini-batch gradient
(1)
neural network
(1)
spectral norm regularization
(1)
network growth
(1)
Papers
Deep Fusion: Efficient Network Training via Pre-trained Initializations
ICML 2024
The Impact of Geometric Complexity on Neural Collapse in Transfer Learning
NeurIPS 2024
Why neural networks find simple solutions: The many regularizers of geometric complexity
NeurIPS 2022
Discretization Drift in Two-Player Games
ICML 2021
Implicit Gradient Regularization
ICLR 2021
On the Origin of Implicit Regularization in Stochastic Gradient Descent
ICLR 2021