Picture for Dong-Young Lim

Dong-Young Lim

Dual Cone Gradient Descent for Training Physics-Informed Neural Networks

Add code
Sep 27, 2024
Viaarxiv icon

On diffusion-based generative models and their error bounds: The log-concave case with full convergence estimates

Add code
Nov 22, 2023
Viaarxiv icon

Langevin dynamics based algorithm e-TH$\varepsilon$O POULA for stochastic optimization problems with discontinuous stochastic gradient

Add code
Oct 24, 2022
Viaarxiv icon

Non-asymptotic estimates for TUSLA algorithm for non-convex learning with applications to neural networks with ReLU activation function

Add code
Jul 19, 2021
Figure 1 for Non-asymptotic estimates for TUSLA algorithm for non-convex learning with applications to neural networks with ReLU activation function
Figure 2 for Non-asymptotic estimates for TUSLA algorithm for non-convex learning with applications to neural networks with ReLU activation function
Figure 3 for Non-asymptotic estimates for TUSLA algorithm for non-convex learning with applications to neural networks with ReLU activation function
Figure 4 for Non-asymptotic estimates for TUSLA algorithm for non-convex learning with applications to neural networks with ReLU activation function
Viaarxiv icon

Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks

Add code
May 28, 2021
Figure 1 for Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
Figure 2 for Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
Figure 3 for Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
Figure 4 for Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
Viaarxiv icon