Jaehong Yoon
jaehong.yoon [at] kaist [dot] ac [dot] kr
News
02. 2023
I'm currently on the job market for the
Postdoc or Research Scientist position.
I'm eager to build sustainable learning algorithms for real-world problems in various research fields,
such as continual learning, online learning, federated learning, self-supervised learning, video understanding,
multimodal learning, large language models, open-world problems, reinforcement learning, etc.
Please feel free to reach out.
Please feel free to reach out.
01. 2023
One paper got accepted to
ICLR 2023.
12. 2022
Jaehong successfully
defended his Ph.D. entitled 'On-device, Online Continual Learning for the Real World'!
08. 2022
One paper got accepted to
ECCV 2022 Workshop on Computational Aspects of Deep Learning (CADL).
05. 2022
Two papers got accepted to
ICML 2022.
01. 2022
Two papers got accepted to
ICLR 2022
including one
oral presentation (acceptance rate=54/3391=1.6%).
Introduction
I’m a Ph.D. candidate in the
Machine Learning and Artificial Intelligence (MLAI)
Lab at KAIST, under the supervision of Prof. Sung Ju Hwang.
I finished the B.S. and M.S. degree at UNIST in 2016 and 2018, respectively.
I interned at Microsoft Research (mentor: Dr. Yue Cao) and
worked as a short-term visiting student at Weizmann Institute of Science (host: Prof. Yonina Eldar).
My research interest mainly focuses on developing lifelong-evolving and efficient deep learning alrogithms for deploying sustainable on-device artificial general intelligence systems. In particular, I've been focusing on tackling practical and real-world challenges in application domains, such as online/streaming learning, egocentric videos, and audio-video-text multimodal problems.
New Preprints
[P6] Text-Guided Token Selection for Text-to-Image Synthesis with Token-based Diffusion Models
Preprint, 2023Paper BibTeX
[P5] Continual Learners are Incremental Model Generalizers
Preprint, 2023Paper BibTeX
[P4] Efficient Video Representation Learning via Masked Video Modeling with Motion-centric Token Selection
arXiv:2211.10636, 2022Paper Code
@article{hwang2022efficient,
title={Efficient Video Representation Learning via Masked Video Modeling with Motion-centric Token Selection},
author={Hwang, Sunil and Yoon, Jaehong and Lee, Youngwan and Hwang, Sung Ju},
journal={arXiv preprint arXiv:2211.10636},
year={2022},
}
title={Efficient Video Representation Learning via Masked Video Modeling with Motion-centric Token Selection},
author={Hwang, Sunil and Yoon, Jaehong and Lee, Youngwan and Hwang, Sung Ju},
journal={arXiv preprint arXiv:2211.10636},
year={2022},
}
[P3] Personalized Subgraph Federated Learning
arXiv:2206.10206, 2022Paper
@article{baek2022personalized,
title={Personalized Subgraph Federated Learning},
author={Baek, Jinheon and Jeong, Wonyong and Jin, Jiongdao and Yoon, Jaehong and Hwang, Sung Ju},
journal={arXiv preprint arXiv:2206.10206},
year={2022},
}
title={Personalized Subgraph Federated Learning},
author={Baek, Jinheon and Jeong, Wonyong and Jin, Jiongdao and Yoon, Jaehong and Hwang, Sung Ju},
journal={arXiv preprint arXiv:2206.10206},
year={2022},
}
Publications
[C10] On the Soft-Subnetwork for Few-shot Class Incremental Learning
ICLR 2023Paper Code
@inproceedings{kang2023on,
title={On the Soft-Subnetwork for Few-shot Class Incremental Learning},
author={Kang, Haeyong and Yoon, Jaehong and Madjid, Sultan Rizky Hikmawan and Hwang, Sung Ju and Yoo, Chang D},
booktitle={International Conference on Learning Representation},
year={2023},
url={https://openreview.net/forum?id=z57WK5lGeHd}
}
title={On the Soft-Subnetwork for Few-shot Class Incremental Learning},
author={Kang, Haeyong and Yoon, Jaehong and Madjid, Sultan Rizky Hikmawan and Hwang, Sung Ju and Yoo, Chang D},
booktitle={International Conference on Learning Representation},
year={2023},
url={https://openreview.net/forum?id=z57WK5lGeHd}
}
[W1] BiTAT: Neural Network Binarization with Task-dependent Aggregated Transformation
ECCV 2022 Workshop on Computational Aspects of Deep Learning (CADL)
Paper
@article{park2022bitat,
title={BiTAT: Neural Network Binarization with Task-dependent Aggregated Transformation},
author={Park, Geon and Yoon, Jaehong and Zhang, Haiyang and Zhang, Xing and Hwang, Sung Ju and Eldar, Yonina C},
journal={arXiv preprint arXiv:2207.01394},
year={2022},
}
title={BiTAT: Neural Network Binarization with Task-dependent Aggregated Transformation},
author={Park, Geon and Yoon, Jaehong and Zhang, Haiyang and Zhang, Xing and Hwang, Sung Ju and Eldar, Yonina C},
journal={arXiv preprint arXiv:2207.01394},
year={2022},
}
[C9] Bitwidth Heterogeneous Federated Learning with Progressive Weight Dequantization
ICML 2022Paper Code
@inproceedings{yoon2022bitwidth,
title={Bitwidth Heterogeneous Federated Learning with Progressive Weight Dequantization},
author={Yoon, Jaehong and Park, Geon and Jeong, Wonyong and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={25552--25565},
year={2022},
organization={PMLR}
}
title={Bitwidth Heterogeneous Federated Learning with Progressive Weight Dequantization},
author={Yoon, Jaehong and Park, Geon and Jeong, Wonyong and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={25552--25565},
year={2022},
organization={PMLR}
}
[C8] Forget-free Continual Learning with Winning Subnetworks
ICML 2022Paper Code
@inproceedings{kang2022forget,
title={Forget-free Continual Learning with Winning Subnetworks},
author={Kang, Haeyong and Mina, Rusty John Lloyd and Madjid, Sultan Rizky Hikmawan and Yoon, Jaehong and Hasegawa-Johnson, Mark and Hwang, Sung Ju and Yoo, Chang D},
booktitle={International Conference on Machine Learning},
pages={10734--10750},
year={2022},
organization={PMLR}
}
title={Forget-free Continual Learning with Winning Subnetworks},
author={Kang, Haeyong and Mina, Rusty John Lloyd and Madjid, Sultan Rizky Hikmawan and Yoon, Jaehong and Hasegawa-Johnson, Mark and Hwang, Sung Ju and Yoo, Chang D},
booktitle={International Conference on Machine Learning},
pages={10734--10750},
year={2022},
organization={PMLR}
}
[C7] Representational Continuity for Unsupervised Continual Learning
ICLR 2022 Oral Presentation (Acceptance Rate=1.6%)Paper Code
@inproceedings{madaan2022rethinking,
title={Rethinking the Representational Continuity: Towards Unsupervised Continual Learning},
author={Madaan, Divyam and Yoon, Jaehong and Li, Yuanchun and Liu, Yunxin and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=9Hrka5PA7LW}
}
title={Rethinking the Representational Continuity: Towards Unsupervised Continual Learning},
author={Madaan, Divyam and Yoon, Jaehong and Li, Yuanchun and Liu, Yunxin and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=9Hrka5PA7LW}
}
[C6] Online Coreset Selection for Rehearsal-based Continual Learning
ICLR 2022
Paper Code
@inproceedings{yoon2022online,
title={Online Coreset Selection for Rehearsal-based Continual Learning},
author={Yoon, Jaehong and Madaan, Divyam and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=f9D-5WNG4Nv}
}
title={Online Coreset Selection for Rehearsal-based Continual Learning},
author={Yoon, Jaehong and Madaan, Divyam and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2022},
url={https://openreview.net/forum?id=f9D-5WNG4Nv}
}
[C5] Federated Continual Learning with Weighted Inter-client Transfer
ICML 2020 Workshop on Lifelong Machine LearningICML 2021
Paper Code
@inproceedings{yoon2021federated,
title={Federated continual learning with weighted inter-client transfer},
author={Yoon, Jaehong and Jeong, Wonyong and Lee, Giwoong and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={12073--12086},
year={2021},
organization={PMLR}
}
title={Federated continual learning with weighted inter-client transfer},
author={Yoon, Jaehong and Jeong, Wonyong and Lee, Giwoong and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={12073--12086},
year={2021},
organization={PMLR}
}
[C4] Federated Semi-supervised Learning with Inter-Client Consistency & Disjoint Learning
ICML 2020 Workshop on Federated Learning (Long Presentation) (Best Student Paper Award)ICLR 2021
Paper Code
@inproceedings{jeong2021federated,
title={Federated Semi-Supervised Learning with Inter-Client Consistency \& Disjoint Learning},
author={Jeong, Wonyong and Yoon, Jaehong and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=ce6CFXBh30h}
}
title={Federated Semi-Supervised Learning with Inter-Client Consistency \& Disjoint Learning},
author={Jeong, Wonyong and Yoon, Jaehong and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=ce6CFXBh30h}
}
[C3] Scalable and Order-robust Continual Learning with Additive Parameter Decomposition
ICLR 2020
Paper Code
@inproceedings{Yoon2020Scalable,
title={Scalable and Order-robust Continual Learning with Additive Parameter Decomposition},
author={Yoon, Jaehong and Kim, Saehoon and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=r1gdj2EKPB}
}
title={Scalable and Order-robust Continual Learning with Additive Parameter Decomposition},
author={Yoon, Jaehong and Kim, Saehoon and Yang, Eunho and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2020},
url={https://openreview.net/forum?id=r1gdj2EKPB}
}
[C2] Lifelong Learning with Dynamically Expandable Networks
ICLR 2018
Paper Code
@inproceedings{yoon2018lifelong,
title={Lifelong Learning with Dynamically Expandable Networks},
author={Yoon, Jaehong and Yang, Eunho and Lee, Jeongtae and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2018},
url={https://openreview.net/forum?id=Sk7KsfW0-},
}
title={Lifelong Learning with Dynamically Expandable Networks},
author={Yoon, Jaehong and Yang, Eunho and Lee, Jeongtae and Hwang, Sung Ju},
booktitle={International Conference on Learning Representations},
year={2018},
url={https://openreview.net/forum?id=Sk7KsfW0-},
}
[C1] Combined Group and Exclusive Sparsity for Deep Neural Networks
ICML 2017
Paper Code
@inproceedings{yoon2017combined,
title={Combined group and exclusive sparsity for deep neural networks},
author={Yoon, Jaehong and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={3958--3966},
year={2017},
organization={PMLR}
}
title={Combined group and exclusive sparsity for deep neural networks},
author={Yoon, Jaehong and Hwang, Sung Ju},
booktitle={International Conference on Machine Learning},
pages={3958--3966},
year={2017},
organization={PMLR}
}
[P2] Rapid Structural Pruning of Neural Networks with Set-based Task-Adaptive Meta-Pruning
arXiv:2006.12130, 2020Paper
@article{song2020rapid,
title={Rapid Structural Pruning of Neural Networks with Set-based Task-Adaptive Meta-Pruning},
author={Song, Minyoung and Yoon, Jaehong and Yang, Eunho and Hwang, Sung Ju},
journal={CoRR},
volume={abs/2006.12139},
eprint={2006.12139},
archivePrefix={arXiv},
year={2020},
}
title={Rapid Structural Pruning of Neural Networks with Set-based Task-Adaptive Meta-Pruning},
author={Song, Minyoung and Yoon, Jaehong and Yang, Eunho and Hwang, Sung Ju},
journal={CoRR},
volume={abs/2006.12139},
eprint={2006.12139},
archivePrefix={arXiv},
year={2020},
}
[P1] Adaptive Network Sparsification with Dependent Variational Beta-Bernoulli Dropout
arXiv:1805.10896, 2018Paper
@article{lee2019adaptive,
title={Adaptive Network Sparsification with Dependent Variational Beta-Bernoulli Dropout},
author={Lee, Juho and Kim, Saehoon and Yoon, Jaehong and Lee, Hae Beom and Yang, Eunho and Hwang, Sung Ju},
journal={CoRR},
volume={abs/1805.10896},
eprint={1805.10896},
archivePrefix={arXiv},
year={2018},
}
title={Adaptive Network Sparsification with Dependent Variational Beta-Bernoulli Dropout},
author={Lee, Juho and Kim, Saehoon and Yoon, Jaehong and Lee, Hae Beom and Yang, Eunho and Hwang, Sung Ju},
journal={CoRR},
volume={abs/1805.10896},
eprint={1805.10896},
archivePrefix={arXiv},
year={2018},
}