Kimin Lee

I'm a Ph.D. student in School of Electrical Engineering at Korea Advanced Institute of Science and Technology (KAIST). I'm studying the machine/deep learning during Ph.D (advised by Prof. Jinwoo Shin). During my PhD, I'm also collaborating closely with Prof. Honglak Lee at University of Michigan/Google Brain. Prior to that, I studied the optimization on communication networks during MS degree in School of Electrical Engineering from KAIST.

Contact

  • Email: kiminlee at kaist dot ac dot kr // pokaxpoka at gmail dot com
  • Office: Room 917, N1 IT Building, KAIST, Daejeon
  • Phone: +82-42-350-7632

Research Interests

  • I'm interested in the fields of machine learning and deep learning. Here are my research topics and keywords:
    • Reliable machine/deep learning: Predictive uncertainty, detecting out-of-distribution samples (i.e., novelty detection).
    • Ensemble learning: multiple choice learning, independent ensemble.
    • Adversarial attacks: defending adversarial attacks
    • Stochastic models: Bayesian neural networks and generative models (GAN and VAE).

Updates

  • (2018/09)
    • Our paper was accepted to NIPS 2018 as spotlight presentation
      • Kimin Lee, Kibok Lee, Honglak Lee and Jinwoo Shin, "A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks".
    • Our paper was accepted to NIPS 2018
      • Jonghwan Mun, Kimin Lee, Jinwoo Shin and Bohyung Han, "Learning to Specialize with Knowledge Distillation for Visual Question Answering".
  • (2018/06)
  • (2018/05)
  • (2018/02)
    • Our paper was accepted to CVPR 2018
      • Kibok Lee, Kimin Lee, Kyle Min, Yuting Zhang, Jinwoo Shin and Honglak Lee, "Hierarchical Novelty Detection for Visual Object Recognition".
  • (2018/01)
    • Our paper was accepted to ICLR 2018
      • Kimin Lee, Honglak Lee, Kibok Lee and Jinwoo Shin, "Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples".
    • Invited to present a poster as 2nd Physics Informed Machine Learning