Soo Kyung Kim (김수경), Ph.D

Me in front of LLNL west gate after CCMS program

Contact

Lawrence Livermore National Laboratory. 7000 East Ave, Livermore, CA 94550, USA
(+1) 646-812-5823

jh502125@gmail.com, kim79@llnl.gov,sookyungkim@lbl.gov

[LinkedIn] [CV] [Google Scholar]

Research Interest

Machine Learning Application for Physical Science, Neural Net based Model, Computer Vision, Spatiotemporal data, Reinforcement Learning

Education

Georgia Institute of Technology, Atlanta, Georgia, USA May 2017

Ph.D. in Computational Material Science (Advisor: Prof. Hamid Garmestani)

M. S. in Computational Science & Engineering (Advisor: Prof. Richard Fujimoto)

Thesis: Hybrid Computational Modeling of Thermomagnetic Material Systems

Columbia University, New York, New York, USA May 2009

M. S. in Electrical Engineering

Ewha Woman's University, Seoul, Korea Jun 2007

B. S. in Electrical Engineering, Physics

Summa Cum Laude

Employment

Lawrence Livermore National Lab., Livermore, CA, USA             Jan 2017 - Present

Postdoc Researcher, Center for Applied Scientific Computing

Earth Science and Grid Federation group           

 Supervisor: Dr. Dean N. Williams

·      Tracking and Predicting hurricane in spatio-temporal climate data change using deep learning models elaborating Computer Vision Techniques.

·       Pursuing Deep Learning for Climate Project affiliated with Lawrence Berkeley National Lab.

Material Informatics (LDRD: Internal Research Fund)

                                                                                    Supervisor: Dr. Yong Han

·      Predicting density of molecule using Gated Graphical Neural Networks.

·      Predicting 3D geometry of crystal structure of High Energy Molecule using policy-based Reinforcement Learning

ATOM project

Supervisor: Dr. Jonathan Allen

·      Predicting kinase property of drug molecules based on SMILES features using neural collaborative filtering & conventional matrix factorization methods

·      Design of drug molecules using ML based methods (generative models, auto-encoder, GAN, RL)

Explainable AI using Deep Symbolic Policy (LDRD: Internal Research Fund)

Supervisor: Dr. Brenden Petersen

·      Train RL policy as symbolic equation using RNNs and Deep Symbolic Regression

3-D CT reconstruction (LDRD: Internal Research Fund)

Supervisor: Dr. Kyle Champley

·      Enhance resolution of medical CT image using neural net based super resolution techniques

·      Developing deep learning framework to reconstruct 3D geometry of CT images from limited view of 2D images

 

Sandia National Lab., Livermore, CA, USA          Jan 2016 - Nov 2016

Research Scientist Intern, Hydrogen and Materials Science Department

(Supervisor: Dr. Jonathan Zimmerman, Dr. Catalin Spataru)

·      Developing Monte-Carlo software based on LSF spin model in C++. Analyzing data from ab-initio DFT using machine learning techniques.

·      Studying high temperature Spin-coupling e_ect to statcking fault energy in stainless steel.

 

Lawrence Livermore National Lab, Livermore, CA, USA          Jun - Aug 2014, Jun - Dec 2015

Research Scientist Intern { CCMS program, Physics and Lifescience Division

(Supervisor: Dr. Lorin Benedict, Dr. Mike Surh)

·      Developing Monte-Carlo software based on Heisenberg model in C++, statistically simulating spin thermo-dynamics of FeCoxB(1-x) and CoPt.

·      GPU-utilized parallelization of the Heisenberg Monte-Carlo software.

 

Pacific Northwest National Lab., Richland, WA, USA     Oct 2011 - Dec 2012

Research Student Intern, Advanced Computing, Mathematics and Data Division

(Supervisor: Dr. Kim Ferris)

·      Computing thermo-magnetic properties for MnBi/MnSb using ab-initio MD (NWChem) and Abinitio DFT.

·      Constructing a solvent based carbon capture materials database using SQL

Project

·     Deep Accelerated Science

o  Climate Informatics

§  Detection and Localization of Extreme Climate Events

§  Deep Strom tracking

§  Deep Strom prediction

§  Deep Toxic Micro Dust Prediction in Korea and Japan

§  Super Resolution of Climate modeled output

o  Material Informatics

§  Crystal Structure Prediction of High Energy Material with Reinforcement Learning

·    Distributed Monte-Carlo simulation for magnetic material (Ph.D thesis)

 

Recording of External Presentations

·      Women in Computer Vision at ECCV 2018, Munich, Germany, Tracking and Forecasting extreme climate events using computer vision techniques, Sep 2018


    

·      5 min oral presentation in WACV2019: Deep-Hurricane-Tracker: Tracking and Predicting Extreme Climate Events using ConvLSTM


    

 

Open Source Software

 

·      [HMCS] Heisenberg Monte-Carlo Simulator for Spin Dynamics:
GPU-utilized parallel and distributed simulation method for Monte-Carlo many particle simulation (written in C++ and CUDA)

·      [Deep-Hurricane-Tracker]:
Track and forecasting hurricane using ConvLSTM (written in python and Tensorflow)

Workshop Organization

 

·      Data Mining on Earth System Science (DMESS) at ICDM 2018
Participated as co-organizer and presented invited talk 
https://www.climatemodeling.org/workshops/dmess2018/

·      Big Data in the Geosciences: New Approaches to Storage, Sharing, and Analysis at AGU 2018, Participated as Session Chair

https://agu.confex.com/agu/fm18/meetingapp.cgi/Session/60507/

Publications

  1. Sookyung Kim, Joanne Kim, Brenden K Peterson. An Interactive Visualization Platform for Deep Symbolic Regression. IJCAI - demonstration track (submitted), 2020.
  2. Sookyung Kim, Sunghyun Park, Kangyeol Kim, Joonseok Lee, Junsoo Lee, Jiwoo Lee, Jaegul Choo. Hurricane Nowcasting with Irregular Time-step using Neural-ODE and Video Prediction. International Conference on Learning Representation (ICLR), Tackling Climate Change with Machine Learning (spotlight talk), 2020.
  3. Bhavya Kailkhura, Brian Gallagher, Sookyung Kim, Anna Hiszpanski, T Yong-Jin Han. Reliable and explainable machine-learning methods for accelerated material discovery. Nature, NPJ. Computational Material, 2019.
  4. Sookyung Kim, Yongwoo Cho, Peggy Li, Mike Surh, Yong Han. Physics-guided Reinforcement Learning for 3D Molecular Structures. ML for Physical Science on Neurips, 2019.
  5. Sookyung Kim, Peggy Li, Joanne Kim, Piyush Karende, Yong Han. Optimizing 3D structure of H2O molecule using DDPQ. International Conference on Machine Learning (ICML), RL for Real Life Workshop, 2019.
  6. Sookyung Kim, Sunghyun Park, Sunghyo-Chung, Yun-sung Lee, Joonseok Lee, Hyojin Kim, Mr Parbhat, Jaegul Choo. Focus and Track: Tracking hurricane events. International Conference on Machine Learning (ICML), Climate Change Workshop: How AI can help, 2019.
  7. Sookyung Kim, Sunghyun Park, Sunghyo-Chung, Yun-sung Lee, Joonseok Lee, Hyojin Kim, Mr Parbhat, Jaegul Choo. Focus and Track: Tracking hurricane events. 30th British Machine Vision Conference (BMVC) (spotlight talk), 2019.
  8. Sookyung Kim, Hyojin Kim, Sangwoong Yoon, Joonseok Lee, Samira Kahou, Karthik Kashinath, and Mr Prabhat. Deep-Hurricane-Tracker: Tracking and Predicting Extreme Climate Events using ConvLSTM,WACV(Accepted), 2019.
  9. Sookyung Kim, Sangwoong Yoon, Hyojin Kim, Samira Kahou, Karthik Kashinath, and Mr Prabhat. Tracking Extreme Climate Events in Spatio-temporal Climate Data, European Conference on Computer Vision (ECCV) WiCV Workshop, 2018.
  10. Sookyung Kim, Sangwoong Yoon, Hyojin Kim, Samira Kahou, Karthik Kashinath, and Mr Prabhat. Tracking extreme climate events using Neural Networks, Climate Informatics (with spotlight oral presentation), 2018.
  11. Sookyung Kim, Jungmin. M. Lee, Jiwoo Lee, and Jihoon Seo. Deep-dust: Predicting concentrations of fine dust in Seoul using LSTM, Climate Informatics, 2018.
  12. Sookyung Kim, Mayur Mudigonda, Ankur Mahesh, Samira Kahou, Karthik Kashinath, Dean Williams, Vincent Michalski, Travis O'Brien, Mr Prabhat. Segmenting and Tracking Extreme Climate Events using Neural Networks, Advances in Neural Information Processing Systems (NIPS), Workshop on Deep Learning for Physical Science, 2017.
  13. Sookyung Kim, Sasha Ames, Jiwoo Lee, Chengzhu Zhang, Aaron C. Wilson and Dean Williams. Framework for Detection and Localization of Extreme Climate Event with Pixel Recursive Super Resolution, International Conference on Data Mining (ICDM), Workshop on Data Mining on Earth System Science, 2017.
  14. Sookyung Kim, Sasha Ames, Jiwoo Lee, Chengzhu Zhang,Aaron C.Wilson and Dean Williams. Massive Scale Deep Learning For Detecting Extreme Climate Events, Climate Informatics, 2017.
  15. Sookyung Kim, Machine Learning for Earth System Grid Federation (ESGF), ESGF Proposal Review Meeting, 2017.
  16. Sookyung Kim, Massive Scale Deep Learning for Predicting Extreme Climate Events, Uncertainty Quantification and Data-Driven Modeling, 2017.
  17. Sookyung Kim, Robert Lee, Richard Fujimoto. GPU-Accelerated Heisenberg Monte-Carlo Simulation, The 56th Sanibel Symposium on Quantum Chemistry, Dynamics, Condensed Matter Physics, 2016.
  18. Joonseok Lee, Kisung Lee, Jennifer G. Kim, Sookyung Kim. Personalized Academic Research Paper Recommendation System, ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Workshop on Social Recommender Systems, 2015.
  19. Markus Dane, Lorin Benedict, Mike Surh, Sookyung Kim. Density functional theory calculations of magneto-crystalline anisotropy energies for (Fe1-xCox)2B, Journal of Physics: Condensed Matter, 2014.
  20. Sookyung Kim, Kim Ferris, Dongsheng Li. Thermo-magnetic Properties of Rare-earth Replacement Critical Magnetic Materials from DFT Calculation, International Journal of Computational Material Science, Eng. 01, 1250036, 2012.
  21. Sookyung Kim. Thermo-magnetic Properties of Rare-earth Replacement Magnetic Materials: MnBi/Sb Compounds and MnBi/Sb - Co - Fe alloys, International Conference on Materials Research Society, Document Number 1761613, Paper Number U 3.24, 2013.
  22. Sookyung Kim, Kim Ferris. Thermo-magnetic Properties of Rare-earth Replacement Magnetic Materials, The 52nd Sanibel Symposium on Quantum Chemistry, Dynamics, Condensed Matter Physics, 2012.

Invited Talks

·      ESGF Face to Face meeting, Washington D.C, USA, Deep Learning Application for Analyzing Spatiotemporal Climate Simulation data, Dec 2018

·      Data Mining on Earth System Science at ICDM 2018, Singapore, Singapore, Tracking hurricane events using ConvLSTM, Nov 2018

·      Women in Computer Vision at ECCV 2018, Munich, Germany, Tracking and Forecasting extreme climate events using computer vision techniques, Sep 2018

·      Climate Informatics, Boulder CO, USA, Deep Hurricane Tracker (Spotlight Talk). Sep 2018

·      ESGF Face to Face meeting, San Francisco CA, USA, Deep Learning Application for Community Machine Learning using ESGF Dec 2017

·      Data Mining on Earth System Science, New Orleans L.A, USA, Framework for Detection and Localization of Extreme Climate Event with Pixel Recursive Super Resolution Dec 2017

·      KOCSEA (The Korean Computer Scientists and Engineers Association in America) Symposium, Las Vegas NV, USA, Deep Learning Application for Climate Science Nov 2017

·      Data Analytics Group Seminar in Center for Applied Scientific Computing at LLNL, Livermore CA, USA, Detection, Localization and Recursive Super Resolution of Climate Data

·      Using Deep Learning, Oct 2017

·      ESGF Proposal Review Meeting, Washington D.C., USA, Machine Learning for Earth System Grid Federation (ESGF) Jun 2017

·      Uncertainty Quantification and Data-Driven Modeling, Austin, TX, USA, Massive Scale Deep Learning for Predicting Extreme Climate Events Mar 2017

·      US-Korea Conference (UKC), Dallas, TX, USA, Monte-Carlo approach for simulate annealing and Applying high performance features using GPU Aug 2016

·      NVIDIA GPU Technology Conference (GTC), San Jose, CA, USA, Quantum Monte-Carlo Simulation implementing GPU Apr 2016

·      The 56th Sanibel Symposium on Quantum Chemistry, Dynamics, Condensed Matter Physics, Grunswick, GA, USA, GPU-Accelerated Heisenberg Monte-Carlo Simulation Feb 2016

·      Workshop on Social Recommender Systems in ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Sydney, Australia, Personalized Academic Research Paper Recommendation System Aug 2015

·      Korea Advanced Institute of Science and Technology (KAIST), Department of Physics, Daejeon, South Korea, Design Rules for Rare-earth Replacement Magnetic Materials: MnBi and MnSb Families Dec 2014

·      Materials Research Society (MRS) Fall Meeting 2014, Boston, MA, USA, Thermo-magnetic Properties of Rare-earth Replacement Magnetic Materials Nov 2014

·      US-Korea Conference (UKC), San Francisco, CA, USA, Computational Modeling of Thermo-magnetic Properties of Materials Aug 2014

·      The 52nd Sanibel Symposium on Quantum Chemistry, Dynamics, Condensed Matter Physics, Grunswick, GA, USA, Thermo-magnetic Properties of MnBi and MnSb Binary Compounds with NiAs Structure

References

·      Dean N. Williams (williams13@llnl.gov)

Distinguished Member of Technical Staff, Center for Applied Scientific Computing, Lawrence

Livermore National Lab (Current Supervisor)

·      Hamid Garmestani (hamid.garmestani@mse.gatech.edu)

Professor, Material Science and Engineering, Georgia Institute of Technology (Ph.D. Advisor)

·      Jonathan Zimmerman (jzimmer@sandia.gov)

Manager, Hydrogen and Materials Science Department, Sandia National Lab.

·      Catalin Spataru (cdspata@sandia.gov)

Research Staff member, Materials Physics Department, Sandia National Lab.

·      Lorin Benedict (benedict5@llnl.gov)

Research Staff member, Physics Division, Lawrence Livermore National Lab.

·      Mike Surh (surh1@llnl.gov)

Research Staff member, Computational Materials Science Group, Lawrence Livermore National Lab.

·      Kim Ferris (kim.ferris@pnnl.gov)

Research Staff member