Scott Pesme
7 papers · 2020–2024 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Cross-Pollinator (8) π§ Keyword Pioneer π Conference Polyglot (3) π Interdisciplinary Bridge πΊοΈ Taxonomy Completionist (12)
π£
Hot Topic Early Bird
π
Keyword Champion
(4)
Conferences
NIPS (5)
AISTATS (1)
ICML (1)
Top co-authors
Keywords
stochastic gradient descent
(5)
diagonal linear network
(4)
gradient flow
(3)
implicit regularization
(2)
implicit bia
(2)
mirror descent
(1)
sparse solution
(1)
linear network
(1)
saddle point
(1)
momentum gradient descent
(1)
continuous-time analysis
(1)
robust regression
(1)
overparameterized regression
(1)
adversarial noise
(1)
convergence diagnostic
(1)
edge of stability
(1)
maximum margin classifier
(1)
stochastic gradient flow
(1)
generalization properties
(1)
continuous time analysis
(1)
Papers
Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks
AISTATS 2024
Implicit Bias of Mirror Flow on Separable Data
NIPS 2024
(S)GD over Diagonal Linear Networks: Implicit bias, Large Stepsizes and Edge of Stability
NIPS 2023
Saddle-to-Saddle Dynamics in Diagonal Linear Networks
NIPS 2023
Implicit Bias of SGD for Diagonal Linear Networks: a Provable Benefit of Stochasticity
NIPS 2021
Online Robust Regression via SGD on the l1 loss
NIPS 2020
On Convergence-Diagnostic based Step Sizes for Stochastic Gradient Descent
ICML 2020