21 citations · 26 across the 7 of their papers we have counts for
7 papers
Controllable Feature Whitening for Hyperparameter-Free Bias Mitigation
Yooshin Cho, Hanbyel Cho, Janghyeon Lee +3
As the use of artificial intelligence rapidly increases, the development of trustworthy artificial intelligence has become important. However, recent studies have shown that deep n…
Disposable Transfer Learning for Selective Source Task Unlearning
Seunghee Koh, Hyounguk Shon, Janghyeon Lee +2
Transfer learning is widely used for training deep neural networks (DNN) for building a powerful representation. Even after the pre-trained model is adapted for the target task, th…
UniCLIP: Unified Framework for Contrastive Language-Image Pre-training
Janghyeon Lee, Jongsuk Kim, Hyounguk Shon +4
Pre-training vision-language models with contrastive objectives has shown promising results that are both scalable to large uncurated datasets and transferable to many downstream a…
DLCFT: Deep Linear Continual Fine-Tuning for General Incremental Learning
Hyounguk Shon, Janghyeon Lee, Seung Hwan Kim +1
Pre-trained representation is one of the key elements in the success of modern deep learning. However, existing works on continual learning methods have mostly focused on learning…
TricubeNet: 2D Kernel-Based Object Representation for Weakly-Occluded Oriented Object Detection
Beomyoung Kim, Janghyeon Lee, Sihaeng Lee +2
We present a novel approach for oriented object detection, named TricubeNet, which localizes oriented objects using visual cues ( heatmap) instead of oriented box offsets re…
Continual Learning with Extended Kronecker-factored Approximate Curvature
Janghyeon Lee, Hyeong Gwon Hong, Donggyu Joo +1
We propose a quadratic penalty method for continual learning of neural networks that contain batch normalization (BN) layers. The Hessian of a loss function represents the curvatur…
Residual Continual Learning
Janghyeon Lee, Donggyu Joo, Hyeong Gwon Hong +1
We propose a novel continual learning method called Residual Continual Learning (ResCL). Our method can prevent the catastrophic forgetting phenomenon in sequential learning of mul…