Publications

(2020). Nimble: Lightweight and Parallel GPU Task Scheduling for Deep Learning. NeurIPS 2020 (Spotlight).

PDF

(2019). Stage-based Hyper-parameter Optimization for Deep Learning. Workshop on Systems for ML at NeurIPS 2019.

PDF

(2019). Speculative Symbolic Graph Execution of Imperative Deep Learning Programs. SIGOPS Operating Systems Review, July 2019.

PDF

(2019). Demonstration of JANUS: Fast and Flexible Deep Learning via Symbolic Graph Execution of Imperative Programs. SysML 2019 Demo.

PDF

(2019). Parallax: Sparsity-aware Data Parallel Training of Deep Neural Networks. EuroSys 2019 (acceptance rate 45 / 207 = 21.7%).

Preprint PDF Code

(2019). JANUS: Fast and Flexible Deep Learning via Symbolic Graph Execution of Imperative Programs. NSDI 2019 (acceptance rate 49 / 332 = 14.7%).

Preprint PDF Slides

(2018). Improving the Expressiveness of Deep Learning Frameworks with Recursion. EuroSys 2018 (acceptance rate 43 / 262 = 16.4%).

Preprint PDF

(2017). Auto-Parallelizing Deep Learning for Multi-machine, Multi-GPU Environments. Workshop on AI Systems SOSP 2017.

PDF

(2015). OS-less Dynamic Binary Instrumentation for Embedded Firmware. IEEE Symposium on Low-Power and High-Speed Chips 2015 (COOL Chips XVIII).

PDF

(2013). (Thesis) Reliable basic block coverage analysis of optimized program using dynamic binary analysis. Seoul National University (Outstanding Master’s Thesis Award, written in Korean).

PDF

Experiences

 
 
 
 
 

Visiting Student Researcher

RISELab, EECS at UC Berkeley

Sep 2019 – Dec 2019 Berkeley, CA, USA
 
 
 
 
 

Engineer

Samsung Research

Aug 2013 – Feb 2017 Seoul, Korea

Talks

Fast and Flexible Systems for Deep Learning

  • [20/02/12] KAIST
  • [20/08/12] SNU CSE AI Summer School

Speculative Symbolic Graph Execution of Imperative Deep Learning Programs [slides]

  • [19/04/03] DAWN Lab, Stanford University
  • [19/03/29] NVIDIA (Endeavor)
  • [19/03/28] Google Brain (Mountain View)
  • [19/03/27] Microsoft Research (Palo Alto)
  • [19/03/26] RISELab, UC Berkeley
  • [19/02/28] Google Brain (Cambridge)

JANUS: Fast and Flexible Deep Learning via Symbolic Graph Execution of Imperative Programs

  • [19/02/27] NSDI 2019 [link]
  • [18/11/29] M1522.001600 Topics in Big Data Analytics, graduate course at SNU