Search

Taehyeon Kim’s Personal Blog

I’m a Ph.D. candidate in the Graduate School of AI at Korea Advanced Institute of Science and Technology (KAIST), advised by Prof. Se-Young Yun, and a member of OSI Lab. During my study, I interned at Qualcomm AI ADAS (Seoul, South Korea, 2021). I received a B.S. in Mathematics from KAIST in 2018.
My research has investigated trustworthy and real-world AI/ML challenges. Specifically, my interests include the optimization for training deep neural networks, automated neural architecture search, automated hyperparameter search, learning with noisy labels, model compression, federated learning, and precipitation nowcasting. My research has been presented at several conferences and organizations.
I'm looking for a research internship/visiting scholar position for Winter 2022 / Spring 2023.
I'm eager to build lifelong and expandable learning algorithms in various research areas, such as neural architecture search, hyperparameter optimization, learning with noisy labels, and federated learning. I can relocate for an on-site internship/visit up to six months. If you are interested, feel free to contact me.
Gallery view
Search
Recent News
Default view
Gallery
Search
Title
Conference and Journal
Author
Keyword
Specification
Link
Under Review
1st author
Precipitation Nowcasting
Benchmark Dataset
Under Review
A Bag of Tricks for Federated Learning in Image Claissfication
Open
Working in Progress
Co-author
Federated Learning
Working In Progress
Not Yet
Supernet Training for Federated Image Classification under System Heterogeneity
Open
Workshop
ICML2022
Under Review
1st author
Federated Learning
AutoML
Image Classification
Architecture Search
International
Top-tier
Oral
Under Review for NeurIPS2022
Revisiting Architecture-aware Knowledge Distillation: Smaller Models and Faster Search
Open
Workshop
ICML2022
1st author
Architecture Search
Bayesian Optimization
Knowledge Distillation
Orthogonality Regularization
International
Top-tier
Revisiting Kernel Orthogonality Regularization: A Study for Convolutional Neural Networks in Image Classification
Open
IEEE Access
1st author
Orthogonality Regularization
International
Journal
Mold into a Graph: Efficient Bayesian Optimization over Mixed-Spaces
Open
Arxiv
Co-author
Hyperparameter Optimization
Graph Neural Network
Working In Progress
Accurate and Fast Federated Learning via Combinatorial Multi-Armed Bandits.
Open
Arxiv
Technical Report
1st author
Federated Learning
Hyperparameter Optimization
Technical Report
FINE Samples for Learning with Noisy Labels
Open
NeurIPS2021
1st author
Noisy Label Classification
Image Classification
SVD
International
Top-tier
Comparing Kullback-Leibler Divergence and Mean Squared Error Loss in Knowledge Distillation
Open
IJCAI2021
1st author
Knowledge Distillation
Noisy Label Classification
Image Classification
International
Top-tier
Adaptive Local Bayesian Optimization Over Multiple Discrete Variables
Open
Workshop
Competition
NeurIPS2020
1st author
AutoML
Hyperparameter Optimization
Bayesian Optimization
Multi-armed Bandit
International
8th rank
Efficient Model for Image Classification With Regularization Tricks
Open
NeurIPS2019
PMLR
Competition
1st author
Orthogonality Regularization
Architecture Search
Augmentation
Model Compression
Image Classification
International
Top-tier
2nd rank
Orthogonal feature regularization : a novel approach for training robust models, Korea Advanced Institute of Science and Technology
Open
Master Thesis
KAIST
1st author
Adversarial Attack
Orthogonality Regularization
Image Classification
Thesis