Bum Jun Kim
3 papers · 2023–2026 · 2 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
Top co-authors
Research topics
Deep Learning
(3)
Computer Vision
(2)
Machine Learning
(2)
Architectures
(2)
Core AI · Computer Vision
(2)
Artificial Intelligence
(1)
Application Areas
(1)
Core AI · Artificial Intelligence
(1)
Learning Types
(1)
Optimization & Theory · Machine Learning
(1)
Optimization & Theory · Deep Learning
(1)
Techniques
(1)
Keywords
vision transformer
(1)
data augmentation
(1)
batch normalization
(1)
spectral analysis
(1)
network optimization
(1)
dropout regularization
(1)
residual network
(1)
transformer training
(1)
dynamic mode decomposition
(1)
koopman operator
(1)
training instability
(1)
positional embedding
(1)
global average pooling
(1)
variance shift
(1)
weight layer
(1)
Papers
Residual Koopman Spectral Profiling for Predicting and Preventing Transformer Training Instability
UAI 2026
Configuring Data Augmentations to Reduce Variance Shift in Positional Embedding of Vision Transformers
AAAI 2025
How to use dropout correctly on residual networks with batch normalization
UAI 2023