Welcome to Yuzhe Yang's Homepage!

Hi there! I am a ~~first~~ ~~second~~ third-year Ph.D. student in Computer Science at MIT CSAIL. I am advised by Prof. Dina Katabi. Before coming to MIT, I recieved my B.S. degree with honors from Peking University.

My research interests lie in machine learning, wireless sensing, and AI for healthcare. I design learning-based models to sense and understand human health contactlessly via wireless signals. I am also broadly interested in robustifying machine learning models against adversaries, data imbalance, etc. I also work on developing provable, structured, and more efficient (deep) reinforcement learning algorithms.

/ / / /

News

[10/2020] Two papers (semi-/self-supervised imbalanced learning & sample-efficient RL) are accepted to NeurIPS 2020.
[09/2020] BodyCompass is covered by MIT News and featured on MIT front page.

[04/2020] My recent research on contactless human sleep sensing has been used to let doctors monitor COVID-19 patients from a distance.

[12/2019] Paper on structured reinforcement learning is accepted to ICLR 2020 as an Oral Presentation (1.8%).

[07/2018] My B.S. thesis won the Best 10 Dissertation Award (10 in 400) in School of EECS at Peking University.
[11/2017] My work on 3D air quality monitoring via UAV sensing won the First Prize (No.1 out of 23 finalists) in IEEE ComSoc Student Competition'17.
[Show more]

Publications [Full List]

(* indicates equal contribution)

Rethinking the Value of Labels for Improving Class-Imbalanced Learning
Yuzhe Yang, Zhi Xu
34^th Conference on Neural Information Processing Systems (NeurIPS 2020)
[Paper] • [Project Page] • [Blog Post] • [arXiv] • [Code] • [Slides] • [Talk] • [BibTeX]

Sample Efficient Reinforcement Learning via Low-Rank Matrix Estimation
Devavrat Shah*, Dogyoon Song*, Zhi Xu*, Yuzhe Yang* (α-β order)
34^th Conference on Neural Information Processing Systems (NeurIPS 2020)
[Paper] • [arXiv] • [BibTeX]

Harnessing Structures for Value-Based Planning and Reinforcement Learning
Yuzhe Yang, Guo Zhang, Zhi Xu, Dina Katabi
8^th International Conference on Learning Representations (ICLR 2020)
Oral Presentation [acceptance rate: 1.8%]
[Paper] • [OpenReview] • [Project Page] • [arXiv] • [Code] • [Slides] • [Talk] • [BibTeX]

When NAS Meets Robustness: In Search of Robust Architectures against Adversarial Attacks
Minghao Guo*, Yuzhe Yang*, Rui Xu, Ziwei Liu, Dahua Lin
IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2020)
[Paper] • [Project Page] • [arXiv] • [Code] • [Slides] • [Poster] • [Talk] • [BibTeX]

BodyCompass: Monitoring Sleep Posture with Wireless Signals
Shichao Yue, Yuzhe Yang, Hao Wang, Hariharan Rahul, Dina Katabi
ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2020)
[Paper] • [Project Page] • [Slides] • [Talk] • [BibTeX] • [MIT News]

ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation
Yuzhe Yang, Guo Zhang, Dina Katabi, Zhi Xu
36^th International Conference on Machine Learning (ICML 2019)
[Paper] • [Project Page] • [arXiv] • [Code] • [Slides] • [Poster] • [Talk] • [BibTeX]

ImgSensingNet: UAV Vision Guided Aerial-Ground Air Quality Sensing System
Yuzhe Yang, Zhiwen Hu, Kaigui Bian, Lingyang Song
IEEE International Conference on Computer Communications (INFOCOM 2019)
[Paper] • [Project Page] • [arXiv] • [Code] • [Video] • [BibTeX]

Real-Time Profiling of Fine-Grained Air Quality Index Distribution using UAV Sensing
Yuzhe Yang, Zijie Zheng, Kaigui Bian, Lingyang Song, Zhu Han
IEEE Internet of Things Journal (IoT-J), Feb. 2018
[Paper] • [arXiv] • [Video] • [BibTeX]
IEEE ComSoc Student Competition'17 Best Project (No.1)

Talks (Selected)

Algorithms & Applications for AI-driven Healthcare
@ Artificial Intelligence Society, Johns Hopkins University. Nov 2020.
Rethinking the Value of Labels for Improving Class-Imbalanced Learning
@ JiaoLab Group, UC Berkeley. Nov 2020.
@ NeurIPS 2020, ~~Vancouver~~ Virtual Conference.
Harnessing Structures for Value-Based Planning and Reinforcement Learning
@ ICLR 2020, ~~Addis Ababa~~ Virtual Conference.
When NAS Meets Robustness: In Search of Robust Architectures against Adversarial Attacks
@ CVPR 2020, ~~Seattle~~ Virtual Conference.
ME-Net: Towards Effective Adversarial Robustness with Matrix Estimation
@ ICML 2019, Long Beach.
AQNet: Fine-Grained 3D Spatio-Temporal Air Quality Monitoring by Aerial-Ground WSN
@ INFOCOM 2018, Hawaii.

Honors and Awards (Selected)

Top Ten Undergraduate Dissertation Award, 10 in 400 in EECS Dept., Peking University, 2018.
First Prize (No.1 out of 23 finalists), IEEE ComSoc Student Competition: Communication Technology Changing the World. 2017.
National Scholarship (highest honor), Top 0.1% among undergraduates nationwide, The Chinese Government, 2017.
HuaWei Scholarship, Top 0.7% among EECS undergraduates, Peking University, 2017.
Outstanding Undergraduate Research Award, Top 1% among undergraduates, Peking University, 2017.

Professional Service

Conference Reviewer: NeurIPS, ICML, ICLR, ICCV, ECCV, IJCAI, UAI
Program Committee Member: IJCAI, UAI, AdvML@ECCV, SecureML@AAAI
Journal Reviewer: IEEE Trans. Vehicular Technology, IEEE Internet of Things Journal

Press

BodyCompass was covered by: MIT News, Engadget, Yahoo, Technology Networks, Sleep Review, TechTimes, SlashGear, TechSpot, Hackster, New Atlas, BGR, and other media outlets.
COVID-19 Health Monitoring was covered by: CSAIL News, Tech Crunch, Engadget, VentureBeat, and other media outlets.

Contact

MIT Computer Science & Artificial Intelligence Lab
32-G928, 32 Vassar St.
Cambridge, MA 02139
yuzhe [at] mit [dot] edu