Picture for Bum Jun Kim

Bum Jun Kim

Stochastic Subsampling With Average Pooling

Add code
Sep 25, 2024
Viaarxiv icon

The Disappearance of Timestep Embedding in Modern Time-Dependent Neural Networks

Add code
May 23, 2024
Viaarxiv icon

Configuring Data Augmentations to Reduce Variance Shift in Positional Embedding of Vision Transformers

Add code
May 23, 2024
Viaarxiv icon

Scale Equalization for Multi-Level Feature Fusion

Add code
Feb 02, 2024
Viaarxiv icon

Analysis of NaN Divergence in Training Monocular Depth Estimation Model

Add code
Nov 07, 2023
Viaarxiv icon

Resolution-Aware Design of Atrous Rates for Semantic Segmentation Networks

Add code
Jul 26, 2023
Viaarxiv icon

Understanding Gaussian Attention Bias of Vision Transformers Using Effective Receptive Fields

Add code
May 08, 2023
Viaarxiv icon

How to Use Dropout Correctly on Residual Networks with Batch Normalization

Add code
Feb 13, 2023
Viaarxiv icon

On the Ideal Number of Groups for Isometric Gradient Propagation

Add code
Feb 07, 2023
Viaarxiv icon

Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks

Add code
May 15, 2022
Figure 1 for Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks
Figure 2 for Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks
Figure 3 for Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks
Figure 4 for Guidelines for the Regularization of Gammas in Batch Normalization for Deep Residual Networks
Viaarxiv icon