Yohan Jung

Yohan Jung

Postdoc, RIKEN AIP

Yohan Jung

[CV] [Google scholar] updated at 10/01/23.
I am a postdoc in ABI team of RIKEN AIP. Before this, I worked as a postdoc in Kim Jaechul Graduate School of Artificial Inteligence from KAIST, advised by Professor Juho Lee. I completed my PhD in Industrial Systems & Engineering from KAIST, advised Professor Jinkyoo Park. Before entering the PhD program, I received a BS in Mathematics from the University of Seoul (UOS) and MS in Industrial Systems & Engineering from KAIST, advised Professor Jinkyoo Park.

Work Experience

  • 2024.07 - present: Postdoc, ABI team, RIKEN AIP
  • 2023.03 - 2024.06: Postdoc, Graduate School of AI, KAIST

Education

  • 2018.03 - 2023.02: PhD, Industrial Systems & Engineering, KAIST
  • 2016.03 - 2018.02: MS, Industrial Systems & Engineering, KAIST
  • 2008.03 - 2016.02: BS, Mathematics, UOS

Research Interest

  • Gaussian process model and Kernel method
  • Probabilistic & Bayesian Deep Learning and its Uncertainty Estimation
  • Approximate Bayesian Inference
  • PAC-Bayes generalization bound

Working Paper

(P: preprint, O: others, *:equal contribution)

[O2] Improved Function Space Variational Inference with Informative Priors

  • Yohan Jung, Juho Lee
  • Under review

[O1] Domain Generalization for Time-Series Classification: Leverage Inductive Bias on Frequency Domain

  • Yohan Jung*, Chuanbo Hua*, Jinkyoo Park
  • Under review

Publication (Peer-Reviewed)

(W: workshop, C: conference, J: journal, O: others, *:equal contribution)

[C4] Bayesian Convolutional Deep Sets with Task-Dependent Stationary Prior

  • Yohan Jung, Jinkyoo Park
  • International Conference on Artificial Intelligence and Statistics (AISTATS), 2023
  • [PDF] [CODE]

[C3] Efficient Approximate Inference for Stationary Kernel on Frequency domain

  • Yohan Jung, Kyungwoo Song, Jinkyoo Park
  • International Conference on Machine Learning (ICML), 2022
  • Extension of workshop paper [W1]
  • [PDF] [CODE] [VIDEO]

[J1] Scalable Inference for Hybrid Bayesian Hidden Markov Model Using Gaussian Process Emission

  • Yohan Jung, Jinkyoo Park
  • Journal of Computational and Graphical Statistics (JCGS), 2022
  • Extension of workshop paper [W2]
  • [PDF] [CODE]

[C2] Implicit Kernel Attention

  • Kyungwoo Song, Yohan Jung, Dongjun Kim, Il-Chul Moon
  • Association for the Advancement of Artificial Intelligence (AAAI), 2021
  • [PDF] [CODE]

[W2] Scalable Hybrid Hidden Markov Model with Gaussian Process Emission for Sequential Time-series Observations

  • Yohan Jung, Jinkyoo Park
  • 3nd Symposium on Advances in Approximate Bayesian Inference (AABI), 2020
  • [PDF] [VIDEO]

[W1] Spectral Mixture Kernel Approximation Using Reparameterized Random Fourier Feature

  • Yohan Jung, Jinkyoo Park
  • 2nd Symposium on Advances in Approximate Bayesian Inference (AABI), 2019
  • [PDF]

[C1] Energy storage control based on user clustering and battery capacity allocation

  • Heechang Ryu, Yohan Jung, Jinkyoo Park
  • 2017 IEEE Power & Energy Society General Meeting, 1-5
  • [PDF]

[O1] 예측모델 (Gaussian Process Regression) 을 통한 에너지저장시스템 운영전략

  • 정요한, 박진규
  • 한국시뮬레이션학회 2017
  • [PDF]