Research Interest

Dataflow Accelerators for Deep Learning
  • Optimizing execution of DNN applications on dataflow accelerators using compilation techniques
Software-controlled Memory Management
  • Compiler optimizations for architectures deploying software-controlled memory
Reliable Computing
  • Mitigating effects of soft errors on memories using software techniques

Experiences

Work Experiences

ETRI (Electronics and Telecommunications Research Institute)
Daejeon, Korea
Research Scientist
2021.02 - present
  • Intermittent Computing: Intermittent kernel, compiler techniques

Global Experiences

Arizona State University
Tempe, AZ, USA
Collaborate researcher
2019.02 - 2019.05
  • Developed a framework to optimize DNN executions on dataflow accelerators
Collaborate researcher
2015.11 - 2016.05
  • Developed compiler optimization techniques for more efficient code management on scratchpad memory
Collaborate researcher
2013.01 - 2013.02
  • Developed system-wide soft error vulnerability evaluation framework based on cycle-accurate system simulator (gem5)
  • Conducted a research on accurate cache vulnerability calculation model and its effective implementation
University of California at Irvine
Irvine, CA, USA
Visiting student
2012.06 - 2012.08
  • Conducted a research on cache soft error vulnerability with various cache configurations and protection techniques

Academic Services

Yonsei University
Seoul, Korea
Lecturer
  • Engineering Information Processing (ENG1108) - Fall 2014
  • Computer I (SCI2001) - Summer 2014
  • Computer I (SCI2001) - Winter 2014
Teaching assistant
  • Engineering Information Processing (ENG1108) - Fall 2018
  • Computer Architecture (CSI3102) - Spring 2017
  • Computer System (CSI2107) - Fall 2013, Fall 2012
  • Logic Circuit Design (CSI2111) - Spring 2013, Spring 2012

Publications

Conference Proceedings

  • dMazeRunner: Optimizing Convolutions on Dataflow Accelerators (invited)
  • Shail Dave, Aviral Shrivastava, Youngbin Kim, Sasikanth Avancha and Kyoungwoo Lee
    IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2020.
  • Static Function Prefetching for Efficient Code Management on Scratchpad Memory
  • Youngbin Kim, Kyoungwoo Lee and Aviral Shrivastava
    2019 IEEE 37th International Conference on Computer Design (ICCD). IEEE, 2019
  • Reducing Code Management Overhead in Software-managed Multicores
  • Jian Cai, Yooseong Kim, Youngbin Kim, Aviral Shrivastava and Kyoungwoo Lee
    Proceedings of the Conference on Design, Automation & Test in Europe (DATE). European Design and Automation Association, 2017.
  • Splitting Functions in Code Management on Scratchpad Memories
  • Youngbin Kim, Jian Cai, Yooseong Kim, Kyoungwoo Lee and Avrial Shrivastava
    Computer-Aided Design (ICCAD), 2016 IEEE/ACM International Conference on. IEEE, 2016.
  • Guidelines to Design Parity Protected Write-back L1 Data Cache
  • Yohan Ko, Reiley Jeyapaul, Youngbin Kim, Kyoungwoo Lee and Avrial Shrivastava
    Design Automation Conference (DAC), 2015 52nd ACM/EDAC/IEEE. IEEE, 2015.
  • Accurate Cache Vulnerability Modeling in Presence of Protection Techniques
  • Yohan Ko, Reiley Jeyapaul, Youngbin Kim, Kyoungwoo Lee and Avrial Shrivastava
    1st International ESWEEK Workshop on Resiliency in Embedded Electronic Systems, 2015.

    Journal Articles

  • dMazeRunner: Executing Perfectly Nested Loops on Dataflow Accelerators
  • Shail Dave, Youngbin Kim, Sasikanth Avancha, Kyoungwoo Lee and Aviral Shrivastava
    ACM Transactions on Embedded Computing Systems (TECS) - Special Issue ESWEEK 2019, CASES 2019, CODES+ISSS 2019 and EMSOFT 2019. Volume 18 Issue 5s, October 2019.
  • A Survey on Software Management Techniques for Scratchpad Memories
  • Youngbin Kim and Kyoungwoo Lee
    Communications of the Korean Institute of Information Scientists and Engineers, 35(10), 2017.10, 46-51.
  • Protecting Caches from Soft Errors: A Microarchitect’s Perspective
  • Yohan Ko, Reiley Jeyapaul, Youngbin Kim, Kyoungwoo Lee and Avrial Shrivastava
    Microarchitect’s Perspective”. ACM Transactions on Embedded Computing Systems (TECS), Volume 16 Issue 4, September 2017.

    Posters

  • Accurate Cache Vulnerability Estimation Based on Word-level Behaviors
  • Yohan Ko, Youngbin Kim and Kyoungwoo Lee
    Proc. of the KIISE Korea Computer Congress, 89-91, 2014.

    Skills

    Research Framework

    TVM (compiler framework for deep learning), LLVM (compiler infrastructure), gem5 (architectural simulator)

    Programming

    Python, C/C++, bash, HTML/CSS, R, PHP, SQL, Javascript, Swift

    Tools

    Git, Pandas, Django, Docker, AWS, Vim, Linux, LaTeX

    Awards & Honors

    2018
    Research Grant for Excellent Achievement , Department of Computer Science, Yonsei University
    2016
    Research Scholarship , Microchip Technology
    2014
    Superior Presentation Winner , Korea Computer Congress
    2012 - 2013
    Competitive Scholarship , Department of Computer Science, Yonsei University