Haekyu Park (박해규)

Graduate Research Assistant

• Georgia Institute of Technology, Atlanta, GA

Data Science Intern

• NVIDIA, Austin, TX
• Mentor: Bartley Richardson, Brad Rees, Joe Eaton
• The work was integrated into NVIDIA tutorial session at KDD 2019.

Undergraduate Research Assiatant

• Seoul National University, Seoul, Republic of Korea

Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations

                  @article{hohman2020summit,
                    title={Summit: Scaling Deep Learning Interpretability by Visualizing Activation and Attribution Summarizations},
                    author={Hohman, Fred and Park, Haekyu and Robinson, Caleb and Chau, Duen Horng},
                    journal={IEEE Transactions on Visualization and Computer Graphics (TVCG)},
                    year={2020},
                    publisher={IEEE}
                    url={https://fredhohman.com/summit/}
                  }
                

Visual Analytics for Interpretability on Deep Neural Networks

MLsploit: A Cloud-Based Framework for Adversarial Machine Learning Research

NeuralDivergence: Exploring and Understanding Neural Networks by Comparing Activation Distributions

                  @inproceedings{park2019neuraldivergence,
                    title={NeuralDivergence: Exploring and Understanding Neural Networks by Comparing Activation Distributions},
                    author={Park, Haekyu and Hohman, Fred and Chau, Duen Horng},
                    booktitle={Poster, Pacific Visualization Symposium (PacificVis)},
                    year={2019},
                    publisher={IEEE},
                    url={http://haekyu.com/neural-divergence/}
                  }
                

SIDE: Representation Learning in Signed Directed Networks

                  @inproceedings{KimPLK18,
                      author    = {Junghwan Kim and
                                   Haekyu Park and
                                   Ji{-}Eun Lee and
                                   U. Kang},
                      title     = {{SIDE:} Representation Learning in Signed Directed Networks},
                      booktitle = {Proceedings of the 2018 World Wide Web Conference on World Wide Web,
                                   {WWW} 2018, Lyon, France, April 23-27, 2018},
                      pages     = {509--518},
                      year      = {2018},
                    }
                

A Comparative Study of Matrix Factorization and Random Walk with Restart in Recommender Systems

                  @inproceedings{conf/bigdataconf/ParkJK17,
                      author    = {Haekyu Park and
                                   Jinhong Jung and
                                   U. Kang},
                      title     = {A comparative study of matrix factorization and random walk with restart
                                   in recommender systems},
                      booktitle = {2017 {IEEE} International Conference on Big Data, BigData 2017, Boston,
                                   MA, USA, December 11-14, 2017},
                      pages     = {756--765},
                      year      = {2017},
                    }
                

Recommender System for Videos on Oksusu Application

• With SK Telecom, Seoul, Republic of Korea
• Keywords: Tensorflow, Feedforward Neural Network, Sequence Embedding, Word Embedding, Approx. k-NN
• [Details]
  We recommend videos to users of Oksusu application. Our model uses deep learning to achieve quality recommendation. Our recommendation engine handles massive data on users' behaviors and heterogeneous information on videos.

A Fast and Cost Efficient Data Compression Algorithm with Shared Virtual Memory in Heterogeneous System Architecture

• For my undergraduate graduation essay
• Keywords: OpenCL, GPGPU, SVM, HSA
• [Details]
  • I used GPGPU and Shared Virtual Memory (SVM) in Heterogeneous System Architecture (HSA).
  • I acquired parallel programming thinking and skills, especially with C++ and OpenCL.
  • We propose fast data deduplication methods that use general purpose computing on graphics processing units (GPGPU) and heterogeneous system architecture (HSA) to efficiently execute the great amount of computations. GPGPU and HSA provide a powerful basis for parallel computing in an easy programmable and efficient way.

Personalized Recommendation for Credit Card Rewards

• With Hyundai Card, Seoul, Republic of Korea
• Keywords: Coupled Matrix Factorization, Time Series Data
• [Details]
  We provide personalized recommendations for credit card rewards to customers. The main algorithm is TCMF (Time Coupled Matrix Factorization). We use various side information of users and items and highly improve the recommendation performance.
• [News article (in Korean)]

Social Recommender System with Graph and Rating Information

• For final project of Probabilistic Graphical Model course
• Keywords: Matrix Factorization, Network Embedding, Social Network, PGM
• [Details]
  We designed and implemented a novel recommendation model leveraging social network by network embedding.