Masahito Ueda
8 papers · 2019–2023 · 3 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Interdisciplinary Bridge πΊοΈ Taxonomy Completionist (10) π Conference Polyglot (3) π§ Keyword Pioneer π Cross-Pollinator (9)
π₯
Unstoppable
(5)
β
The Questioner
Conferences
ICLR (4)
ICML (2)
NIPS (2)
Top co-authors
Keywords
escape rate
(2)
stochastic gradient descent
(2)
neural network optimization
(1)
optimization theory
(1)
selective classification
(1)
stationary distribution
(1)
learning rate
(1)
inductive bia
(1)
hessian eigenvalue
(1)
implicit bia
(1)
stochastic differential equation
(1)
activation function
(1)
saddle point
(1)
flat minimum
(1)
neural network
(1)
periodic function
(1)
function extrapolation
(1)
portfolio theory
(1)
uncertainty quantification
(1)
Papers
What shapes the loss landscape of self supervised learning?
ICLR 2023
Convergent and Efficient Deep Q Learning Algorithm
ICLR 2022
Strength of Minibatch Noise in SGD
ICLR 2022
SGD Can Converge to Local Maxima
ICLR 2022
Power-Law Escape Rate of SGD
ICML 2022
Noise and Fluctuation of Finite Learning Rate Stochastic Gradient Descent
ICML 2021
Neural Networks Fail to Learn Periodic Functions and How to Fix It
NIPS 2020
Deep Gamblers: Learning to Abstain with Portfolio Theory
NIPS 2019