Research

Major Research Achievements

Big Data Analysis Research Laboratory (Professor Park No-sung) Publish…

페이지 정보

profile_image

작성자 최고관리자

댓글 0 comment 조회 316 view 작성일 2022-02-28 19:11

본문

The Big Data Analysis Research Institute (Professor Park No-sung) published a total of 14 papers (12 authors in charge of Professor Park No-sung, 2 co-supervised) at this year's Top Tier Academic Conference in 2021. The research fields that account for a large part of the related research performance are largely the following three areas.


  • ● It shows world-class performance in physics-informed Neural Networks, Neural Ordinal Differential Equations and Neural Controlled Differential Equations, and time series data processing using them.
  • ● Generative Adversarial Networks-like generative models. In particular, it shows the world's best performance in the field of tabular data synthesis.
  • ● It also shows the highest level of performance in various recommendation fields. In particular, regarding recommendations, as a member of Naver's news recommendation algorithm review committee, it is currently verifying that there is no bias or error in Naver's recommendation system.


Among the research results in 2021, the summary and detailed research papers of 14 Toptier papers, including NeurIPS, KDD, ICML, ACL, WWW, and AAAI, are as follows.


Top tier performance : 1 AAAI, 2 SDM, 1 WWW, 1 ACL, 1 ICML, 2 KDD, 2 CIKM, 2 ICDM, 2 NeurIPS


  1. 1. Jungeun Kim, Kookjin Lee, Dongeun Lee, Sheo Yon Jin, Noseong Park, “DPM: A Novel Training Method for Physics-Informed Neural Networks in Extrapolation,” AAAI 2021
    2. Jinsung Jeon, Jing Liu, Jayoung Kim, Jaehoon Lee, Noseong Park, Jamie Jooyeon Lee, Ozlem Uzuner, Sushil Jajodia , "Scalable Graph Synthesis with Adj and 1 — Adj," SDM 2021
    3. Jinsung Jeon, Dongeun Lee, Seunghyun Hwang, Soyoung Kang, Noseong Park, Duanshun Li, Kookjin Lee, Jing Liu, "Large-Scale Flight Frequency Optimization with Global Convergence in the US Domestic Air Passenger Markets," SDM 2021
    4. Jayoung Kim, Jinsung Jeon, Jaehoon Lee, Jihyeon Hyeong, Noseong Park, “OCT-GAN: Neural ODE-based Conditional Tabular GANs,” WWW 2021
    5. Sheo Yon Jhin, Minju Jo, Taeyong Kong, Jinsung Jeon, Noseong Park, “ACE-NODE: Attentive Co-Evolving Neural Ordinary Differential Equations,” KDD 2021
    6. Duanshun Li, Jing Liu, Jinsung Jeon, Seoyoung Hong, Thai Le, Dongwon Lee, Noseong Park "Large-Scale Data-Driven Airline Market Influence Maximization," KDD 2021
    7. Duanshun Li, Jing Liu, Dongeun Lee, Ali Seyedmazloom, Giridhar Kaushik, Kookjin Lee, Noseong Park, "A Novel Method to Solve Neural Knapsack Problems," ICML 2021
    8. Thai Le, Noseong Park, Dongwon Lee, "A Sweet Rabbit Hole by DARCY: Using Honeypots to Detect Universal Trigger’s Adversarial Attacks," ACL 2021
    9. Jeongwhan Choi, Jinsung Jeon, Noseong Park, "LT-OCF: Learnable-Time ODE-based Collaborative Filtering," CIKM 2021
    10. Jinsung Jeon, Soyoung Kang, Minju Jo, Seunghyeon Cho, Noseong Park, Seonghoon Kim, Chiyoung Song, "LightMove: A Lightweight Next-POI Recommendation for Taxicab Rooftop Advertising," CIKM 2021
    11. Sheo Yon Jhin, Heejoo Shin, Seoyoung Hong, Solhee Park, Noseong Park, "Attentive Neural Controlled Differential Equations for Time-series Classification and Forecasting," ICDM 2021
    12. Jeehyun Hwang, Jeongwhan Choi, Hwangyong Choi, Noseong Park, "Climate Modeling with Neural Diffusion Equations," ICDM 2021
    13. Kanghyun Choi, Deokki Hong, Noseong Park, Youngsok Kim, Jinho Lee, “Qimera: Data-free Quantization with Synthetic Boundary Supporting Samples,” NeurIPS 2021
    14. Jaehoon Lee, Jihyeon Hyung, Jinsung Jeon, Noseong Park, “Invertible Tabular GANs: Killing Two Birds with One Stone for Tabular Data Synthesis,” NeurIPS 2021


In particular, 11 Attractive Neural Controlled Differential Equations studies on the list are invited by Professor Terry Lyons' research group at Oxford University in the UK, and Jin Seo-yeon (second semester of Ph.D.) plans to hold a seminar on November 3 with Oxford researchers. The promotional links for the seminar are as follows.


https://www.datasig.ac.uk/event/sheoyon-jhin-2021-11-03


Professor Terry Lyons is a world-renowned scholar who won the Fields Medal of the Nobel Prize in Mathematics for Rough Path Theory and recently proposed groundbreaking methods such as Neural Controlled Differential Equations and Neural Rough Differential Equations in deep learning research.


In addition, Jeon Jin-sung (4th semester of Ph.D.) has produced top tier performance of 4 hesitant and 4 co-authors this year alone, and many studies are currently under review, showing extraordinary research results.

댓글목록

등록된 댓글이 없습니다.