Welcome to Deep Learning & Big Data Systems Lab

In Deep Learning & Big Data Systems Lab, we investigate programmability, scalability,  manageability, and performance of deep learning & big data systems. We study improving existing deep learning systems such as Tensorflow or PyTorch, as well as big data systems such as Hadoop. We investigate various trade-offs (such as performance and accuracy) in existing systems. We also work on security aspect of deep learning models and systems. See Research Page for the (partial) list of research projects.

Research Topics include  (연구 주제):
  • Large-Scale Deep Learning System     대용량 딥 러닝 시스템
  • Deep Learning System Security           딥러닝 시스템 보안
  • Trade-Offs in Deep Learning Model    딥러닝 모델 트레이드 오프 (성능/정확도)
  • Big Data Systems                                빅 데이터 시스템
  • Big Data/Graph Mining and Analysis   빅 데이터/그래프 분석
  자세한 내용은 Research Page 참고

  • NEW: Our paper on neural network quantization is accepted to CC (compiler construction) 2020; CC is a leading conference on compilers and optimizers, where researchers at top-tier schools publish. Congrats to the authors!
  • NEW: Our Lab has two top-tier papers accepted -- VLDB 2019 and Usenix ATC 2019 (in collaboration with KAIST and UNIST). Congrats to the authors!
  • We have openings for MS/PhD students and postdocs in 2018 winter and onward. If you are interested, please contact Prof. Seo (lastname firstname @ hanyang.ac.kr [seojiwon])

My talk at Hadoop Summit ‎‎‎‎‎‎‎‎‎(in collaboration with Hortonworks)‎‎‎‎‎‎‎‎‎ 빅데이터 관련 학회인 하둡 서밋 발표내용 ‎‎(호톤웍스와 함께 발표함)‎‎