Meng-Shiun(Kobe) Yu

Email: Kobeyu.tw at gmail.com

I'm currently an NPU compiler engineer at Kneron, also a Ph.D. student at the department of computer science, National Tsing Hua University, Taiwan. My research interests include 

compiler and runtime for deep learning accelerator;  architecture/performance evaluations; compiler designs and optimizations.

Education


  • Ph.D., Department of Computer Science, National Tsing Hua University                                           2017/9 ~ Present
  • M.A., Department of Electrical Engineering, National Chung Cheng University                                 2006/9 ~ 2008/10
    • Advisor: Huei-Yung Lin
    • Thesis: A Vision Surveillance System for Mobile Robot Using Omni-directional and PTZ Cameras
  • B.A., Department of Electrical Engineering, National Chin Yi University of Technology                    2004/9 ~ 2006/6

Work Experience

Kneron.NPU Compiler Engineer.2019 / 2 ~ Present

  • Participated in the development of 6 NPU SoCs.
  • Developed compiler to compress weights for NPU, metadata for firmware control flow.
  • Propose a whole new NPU executable format(NEF) for the runtime.
  • Developed binary utilities to merge/extract multiple models to/from single NEF file.
  • Developed just in time compiler for image accelerator, used in freeRTOS and Embedded Linux environment.
  • Pre- and post-silicon verification(NPU and image accelerator).
  • Debug mismatch between CSim and RTL or performance issue with firmware, CSim, and HW team.
  • Survey advanced techniques, compiler: TVM, MLIR, runtime: TFRT, IREE, model/hardware co-design: network architecture search.
  • Recruit and lead interns to complete specific research work.

Fongfu.Software Engineering Manager.2016 / 9 ~ 2018 / 6

  • Build up a software team (~5 engineer) to deploy services on iOS/Android and Asus Zenbo.
  • System architecture evaluation and planning (Mobile App,  Web backend)
  • Developed face recognition algorithm.

SmartAll.Senior Software Engineer.2016 / 2 ~ 2016 / 8

  • Developed iOS App to control smart camera.
  • Porting ffmpeg and face detection algorithm on iOS.
  • Developed peer to peer commutation SDK.

MSI.Software Engineer.2013 / 11 ~ 2016 / 2

  • Developed iOS App to control smart camera.

CEO(Create Electronic Optical Co., Ltd).Software Engineer.2010 /1 ~ 2013/3

  • Developed lane departure and forward collision warning systems using computer vision algorithms.
  • Developed media player for driver recorder.

Skills


  • Programming Language
    • C/C++, Python and Shell script
  • Compiler
    • Construction and optimization
  • Deep Learning
    • Optimize and deploy neural network model to edge devices.

Publications

  • The Support of MLIR HLS Adaptor for LLVM IR, Geng-Ming Liang, Chuan-Yue Yuan, Meng-Shiun Yu, Tai-Liang Chen, Kuan-Hsun Chen, and Jenq-Kuen Lee, ICPP EMS 2022, Bordeaux, France, Aug. 29 - Sep. 1, 2022. (Virtual).
  • C++OpenCL4TVM: Support C++OpenCL Kernel for TVM NN Operators, Po-Yao Chang, Tai-Liang Chen, Yu-Tse Huang, Meng-Shiun Yu, Jenq-Kuen Lee, IWOCL 2022 (poster paper), 2022.
  • Optimization with TVM Hybrid OP on RISC-V with P Extension, Chuan-Yue Yuan, Meng-Shiun Yu, Chao-Lin Lee, Chun-Chieh Yang, and Jenq-Kuen Lee, TVMcon 2021, Seattle, Dec. 15-17, 2021
  • Support NNEF Execution Model for NNAPI, Yuan-Ming Chang, Chia-Yu Sung, Yu-Chien Sheu, Meng-Shiun Yu, Min-Yih Hsu, and Jenq-Kuen Lee, Journal of Supercomputing, 77, 10065-10096, Springer, 2021.
  • NNBlocks: A Blockly Framework for AI Computing, Tai-Liang Chen, Yi-Ru Chen, Meng-Shiun Yu, and Jenq-Kuen Lee, Journal of Supercomputing, 77, 8622-8652, Springer, 2021
  • Accelerating NNEF framework on OpenCL devices using clDNN, Meng-Shiun Yu, Tai-Liang Chen and Jenq-Kuen Lee, IWOCL 2020, Germany, May 2020 (poster paper).
  • A Robot Vision System for Visual Surveillance, M. S. Yu, H. Wu, H. Y. Lin, International Journal of Innovative Computing, Information and Control, Vol. 10, No. 4, pp. 1267-1274, Aug. 2014.
  • Lane departure and front collision warning using a single camera, H. Y. Lin, L. Q. Chen, M. S. Yu, International Symposium on Intelligent Signal Processing and Communication System, pp. 64-69, Nov. 2012. 

Talks

Neural Network File Format for Inference Framework.COSCUP.2020/08/02

機器學習應用於蔬果辨識.台灣人工智慧學校 台中分校.2018/11/28 

社群與企業之開源專案合作經驗分享:以阿龜微氣候天眼通為例 .g0v summit.2018/10/6

Agrino應用於農業感測的開源專案.COSCUP.2018/8/11

那些夜裡用啤酒打造的田間感測器.COSCUP.2017/8/6