jaemin_yoo

Jaemin Yoo

Assistant Professor at KAIST

I am an Assistant Professor in the School of Electrical Engineering at KAIST and jointly affiliated with the Kim Jaechul Graduate School of AI. Previously, I was a postdoctoral research fellow at Carnegie Mellon University, where I worked with Prof. Leman Akoglu and Prof. Christos Faloutsos. I received my Ph.D. and B.S. in Computer Science and Engineering from Seoul National University. I received the Google PhD Fellowship, the Qualcomm Innovation Fellowship, and the outstanding dissertation awards from Seoul National University and the Korean Academy of Science and Technology. My research interests cover various topics in data mining and machine learning, including graph neural networks, time series analysis, recommender systems, and anomaly detection, especially based on self-supervised learning with insufficient labels.

News


  • [Oct. 2024] Our work on unsupervised anomaly detection got accepted to NeurIPS 2024.
  • [Aug. 2024] I delivered an invited talk at Korean AI Association about time series forecasting.
  • [Jun. 2024] I delivered an invited talk at KCC 2024 as a new faculty member.
  • [May 2024] Our work on the feature homophily in GNNs got accepted to ICML 2024.
  • [Apr. 2024] I delivered an invited talk at KAIST EE Colloquium about anomaly detection.
  • [Mar. 2024] I delivered an invited talk at Samsung Electronics about recent trends in AI.
  • [Jan. 2024] Our work on network effect analysis got accepted to PAKDD 2024.
  • [Jan. 2024] Our work on self-supervised learning for hypergraphs got accepted to ICLR 2024.
  • [Jan. 2024] I delivered a keynote talk at the ASTAD workshop during WACV 2024.
  • [Oct. 2023] Our vision paper on anomaly detection got accepted to BigData 2023.
  • [Oct. 2023] I delivered a guest lecture on graph augmentation at Yonsei University.
  • About Me


    Positions

    • KAIST (Aug. 2023 - Present)
      Assistant Professor, School of Electrical Engineering
      Adjunct Professor, Kim Jaechul Graduate School of AI
      Research Group: KAIST Data AI Lab
    • Carnegie Mellon University (Mar. 2022 - Jun. 2023)
      Postdoctoral Research Fellow, Heinz College
      Advisors: Prof. Leman Akoglu and Prof. Christos Faloutsos

    Education

    • Seoul National University (Mar. 2016 - Feb. 2022)
      Ph.D. in Computer Science and Engineering
      Advisor: Prof. U Kang
    • Seoul National University (Mar. 2012 - Feb. 2016)
      B.S. in Computer Science and Engineering

    Awards and Honors

    Publications


    2024

    • Rethinking Reconstruction-based Graph-level Anomaly Detectors: Limitations and a Remedy
      Sunwoo Kim, Soo Yong Lee, Fanchen Bu, Shinhwan Kang, Kyungho Kim, Jaemin Yoo, Kijung Shin
      NeurIPS 2024 [ paper | code and datasets | bib ]
    • Feature Distribution on Graph Topology Mediates the Effect of Graph Convolution: Homophily Perspective
      Soo Yong Lee, Sunwoo Kim, Fanchen Bu, Jaemin Yoo, Jiliang Tang, Kijung Shin
      ICML 2024 [ paper | poster | code and datasets | bib ]
    • NetEffect: Discovery and Exploitation of Generalized Network Effects
      Meng-Chieh Lee, Shubhranshu Shekhar, Jaemin Yoo, Christos Faloutsos
      PAKDD 2024 [ paper | code and datasets | bib ]
    • HypeBoy: Generative Self-Supervised Representation Learning on Hypergraphs
      Sunwoo Kim, Shinhwan Kang, Fanchen Bu, Soo Yong Lee, Jaemin Yoo, Kijung Shin
      ICLR 2024 [ paper | poster | code and datasets | bib ]
    • Representative and Back-in-Time Sampling from Real-world Hypergraphs
      Minyoung Choe, Jaemin Yoo, Geon Lee, Woonsung Baek, U Kang, and Kijung Shin
      Transactions on Knowledge Discovery from Data [ paper | code and datasets | bib ]

    2023

    2022

    2021

    • Accurate Graph-Based PU Learning without Class Prior
      Jaemin Yoo*, Junghun Kim*, Hoyoung Yoon*, Geonsoo Kim, Changwon Jang, and U Kang
      ICDM 2021 [ paper | slides | bib | blog (Korean) ]
      One of the best-ranked papers of ICDM 2021 for fast-track journal invitation
    • Accurate Multivariate Stock Movement Prediction via Data-Axis Transformer with Multi-Level Contexts
      Jaemin Yoo, Yejun Soun, Yong-chan Park, and U Kang
      KDD 2021 [ paper | slides | datasets | bib ]
    • Gaussian Soft Decision Trees for Interpretable Feature-Based Classification
      Jaemin Yoo and Lee Sael
      PAKDD 2021 [ paper | slides | code and datasets | bib ]
    • Attention-Based Autoregression for Accurate and Efficient Multivariate Time Series Forecasting
      Jaemin Yoo and U Kang
      SDM 2021 [ paper | slides | bib | blog (Korean) ]

    2020

    • Sampling Subgraphs with Guaranteed Treewidth for Accurate and Efficient Graphical Inference
      Jaemin Yoo, U Kang, Mauro Scanagatta, Giorgio Corani, and Marco Zaffalon
      WSDM 2020 [ paper | poster | code and datasets | bib | blog (Korean) ]
      Samsung HumanTech Paper Award & Qualcomm Innovation Fellowship Korea

    2019

    2017 - 2018

    Miscellaneous


    Teaching

    • EE412: Foundation of Big Data Analytics (Fall 2023; 2024)
    • EE213: Discrete Methods for Electrical Engineering (Spring 2024)

    Professional Services

    Invited Talks


    Jaemin Yoo @ KAIST