About Me

I'm a Ph.D. student at Zhiyuan Honors Program at Shanghai Jiao Tong University , majored in Computer Science. I am a member of APEX Lab, advised by Prof. Weinan Zhang and Prof. Yong Yu.

I am a PMC member of Apache TVM, working closely with Tianqi Chen. I am working on TVM Unity, including TensorIR, Meta-Schedule, Auto-Tensorization and Relax (next Relay).

My research interests include deep learning compiler, machine learning system and urban data computing. I am also interested in distributed systems and computer networks.

Contact Details

Shanghai Jiao Tong University
800 Dongchuan RD. Minhang District
Shanghai, China
Hzfengsy@sjtu.edu.cn

Download Resume

Education

Shanghai Jiao Tong University

Ph.D. in Computer Science • 2020 - Present

I'm a member of Zhiyuan Honors Ph.D. Program

Zhiyuan Honors Program for Ph.D. students is a high-level talent training program aiming to cultivate a group of leading talents in future science and engineering technology with broad vision, scientific spirit and innovative ability

Shanghai Jiao Tong University

B.Sc. in Computer Science • 2016 - 2020

I'm a member of ACM Honors Class, Zhiyuan Collage.

Zhiyuan Collage is a collage for training outstanding students in the basic sciences, while ACM Honors Class is an elite CS program for top 5% talented students.

Research

APEX Data & Knowledge Management Lab

Shanghai Jiao Tong University • July 2018 - Present

  • Member of APEX Lab, advised by Prof. Weinan Zhang and Prof. Yong Yu.
  • Now I am working on TVM, an exciting deep learning compiler stack.
  • My major research topic is deep learning compiler, machine learning system and urban data computing.

SAMPL

University of Washington• July 2019 - Dec 2019

Publications

CityFlow: A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario

WWW 2019 Demo• May 2019 [arXiv]

Huichu Zhang, Siyuan Feng, Chang Liu, Yaoyao Ding, Yichen Zhu, Zihan Zhou, Weinan Zhang, Yong Yu, Haiming Jin, Zhenhui Li

CoT: Cooperative Training for Generative Modeling of Discrete Data

ICML 2019• May 2019 [arXiv]

Sidi Lu, Lantao Yu, Siyuan Feng, Yaoming Zhu, Weinan Zhang, Yong Yu

Awards

Outstanding Graduates

Shanghai Jiao Tong University• 2020

Xing Cai Scholarship

Shanghai Jiao Tong University• 2018

Merit Student

Shanghai Jiao Tong University• 2018

Zhiyuan Honorary Scholarship

Shanghai Jiao Tong University• 2016 - 2018

Zhiyuan Leadership Scholarship

Shanghai Jiao Tong University• 2017

Teaching

CS420: Machine Learning

Shanghai Jiao Tong University • Spring 2019 • Course Page

  • Designed and Judged one of the course projects.
  • One of the judgers for post session.

MS106: Principle and Practice of Computer Algorithms

Shanghai Jiao Tong University • Summer 2018 • Course Page

  • Student Instructor
  • Led the TA team (was one of the three lead TAs).
  • Maintained the Online Judge System for assignments and exams. And built a new judge core for it.
  • Instructed a group of students to implement a simplified Deep Learning System with some simple APIs which is same as those in TensorFlow.

CS147: Data Structures

Shanghai Jiao Tong University • Spring 2018 • Course Page

  • Teaching Assistant
  • Led the TA team (was one of the three lead TAs).
  • Prepared the course project (a railway ticket management system) with other TAs.
  • Prepared the course project (STLite in C++).

CS151: C++ Programming (A)

Shanghai Jiao Tong University • Autumn 2017 • Course Page

  • Teaching Assistant
  • Gave a lecture on C++ multithreading programming.
  • Prepared the course project (Quoridor Game) with other TAs.
  • Prepared the course project (OOP programming in C++) with other TAs.

Apache TVM

Open Deep Learning Compiler Stack

  • Compilation of deep learning models in Keras, MXNet, PyTorch, Tensorflow, CoreML, DarkNet into minimum deployable modules on diverse hardware backends.
  • Infrastructure to automatic generate and optimize tensor operators on more backend with better performance.

CityFlow

A Simulator for Large Scale City Traffic Scenario

  • CityFlow is a new designed open-source traffic simulator, which is much faster than SUMO (Simulation of Urban Mobility).
  • CityFlow can support flexible definitions for road network and traffic flow based on synthetic and real-world data.
  • It also provides user-friendly interface for reinforcement learning.

M-Compiler

A Compiler for M* Language

  • It is a compiler implemented in Java with the source language M*, which is a C-and-Java-like language.
  • Implemented many optimizations for the compiler, which got it faster than gcc -O1.
  • Final score: 95 / 100

KitJudge

A Flexible Online Judge Platform

  • It is a general online judge platform with high degree of freedom, which can be used in most situations.
  • Easy to distribute judge task to different judge clients to speed up the judging process.
  • Carefully designed sandbox system to prevent spiteful code.

RISC-V CPU

A RISC-V CPU on FPGA

  • Designed and implemented a FPGA-supported RISC-V CPU with 5-stage pipeline implemented in Verilog HDL.