Junhong Shen

Junhong Shen

1st-Year Ph.D. Student in Machine Learning

Carnegie Mellon University

I’m a first-year Ph.D. student in the Machine Learning Department at CMU, advised by Ameet Talwalkar. My current research focuses on automating machine learning in real-world applications, in particular democratizing model learning and tuning for diverse tasks outside the widely-studied vision and language domains. Besides, I’m also interested in studying theories of explainable machine learning.

I obtained my B.S. in Mathematics of Computation at UCLA, where I was fortunate to work with Dr. Lin Yang on sample-efficient reinforcement learning. I also studied multi-agent RL and Theory of Mind, advised by Professor Song-Chun Zhu and Professor Ying Nian Wu.

Recent Updates

Interests

  • AutoML and Interpretable ML
  • Reinforcement Learning and Control

Education

  • Doctor of Philosophy, Machine Learning

    Carnegie Mellon University, 2021 - present

  • B.S. in Mathematics of Computation (GPA 4.0/4.0)

    University of California, Los Angeles, 2017 - 2021

Experience

 
 
 
 
 

Research Assistant

Department of Computing and Mathematical Sciences, California Institute of Technology

Oct 2020 – Mar 2021 Los Angeles, CA
Worked with Professor Yisong Yue and Ph.D. student Cameron Voloshin on off-policy policy evaluation.
 
 
 
 
 

Research Assistant

Center for Vision, Cognition, Learning, and Autonomy

Jan 2019 – June 2021 Los Angeles, CA
Work on integrating Theory of Mind into various machine learning settings to improve algorithm efficiency.
 
 
 
 
 

Undergraduate Researcher

Computational Metastasis Lab, Fields Institute for Research in Mathematical Sciences

Jul 2019 – Sep 2019 Toronto, Canada
Studied vascular network reconstruction with clinical images and global constructive optimization.
 
 
 
 
 

Product Manager Intern

Face ID Team, Beijing SenseTime Technology Development Company

Jun 2018 – Sep 2018 Beijing, China
Assisted in the 3D structured light Face ID project; monitored face data collection; collaborated with engineers in software updates, bug fixing, and testing; participated in 5 software version releases.
 
 
 
 
 

Project Lead

Open Project Space Program, UCLA IEEE

Sep 2017 – Jun 2018 Los Angeles, CA
Implemented an autonomous maze-solving robot that avoids obstacles and navigates in a maze.
 
 
 
 
 

Research Assistant

MOE Key Laboratory of Protein Sciences, School of Life Sciences, Tsinghua University

Sep 2015 – Jun 2017 Beijing, China
Researched on the catalytic mechanism of enzyme HPPD.

Projects

FRAME Model and Deep FRAME for Texture Synthesis

Experimenting with the FRAME and Hierarchical FRAME model.

Micromouse: An Autonomous Maze-Solving Robot

Learning how to find goals and avoid obstacles within a maze.

Unlock Your Phone with 3D Structured Light Face ID

Exploring face detection and recognition with deep learning.

Films and Theaters

ColdBlue Film Studio, UCLA

Wrote the script for the documentary On the Road, which was presented on the 2019 UCLA Chinese Cultural Night.

Cfan Chinese Theatre Group, UCLA

Participated in the stage management of three plays, The Enemy of the People, Birthday Party, and The Good Person of Szechwan.