My name is Yue ZHAO (赵越 in Chinese). I am a third-year Ph.D. student at Carnegie Mellon University (CMU), and an ex management consultant at PwC Canada. I have led/participated > 10 ML open-source initiatives, receiving 10,000 GitHub stars (top 0.002%: ranked 900 out of 40M GitHub users) and >400,0000 total downloads. Popular ones:
I specialize in designing and building machine learning systems (MLSys), with realization and applications in outlier detection, healthcare, graph neural networks, and ensemble learning. My research focuses on the intersection of two fields:
At CMU, I work with Prof. Leman Akoglu from DATA Lab on outlier detection, Prof. George H. Chen on general ML and statistics, and Prof. Zhihao Jia from Catalyst on machine learning systems. I am currently visiting Prof. Jure Leskovec at SNAP, Standford University.
Startup and VC: I am interested in capitalizing my expertise in anomaly detection. Let's connect!
General Notes: I am open to ML/DM Internship (2022). Please reach out :)Contact me by Email (zhaoy [AT] cmu.edu) or WeChat (微信) @ yzhao062.
[#1] I am open to collaboration opportunities (anytime & anywhere) and research internships (summer 2022). I could legally work in United States (CPT), Canada (permanent residency), and China (permanent residency). I have been working with the professionals from both industry and academia (e.g., Stanford, Havard, Facebook).
[#2] Call for review oppt. I am looking for paper review, tutorial, workshop, and talk opportunities (in anomaly detection, scalable ML, machine learning systems, and AutoML).
[#3] I host a WeChat group on anomaly detection (异常检测微信讨论组), along with more than three hundred of researchers (e.g., Berkley, MIT, Tsinghua, etc.) and industry people (e.g., Alibaba, IBM, Facebook, etc.) for collaboration and intern/full-time opportunities. Ping me to join!
[#4] I am a dedicated writer with more than 300 articles (in Chinese) and 160,000 followers on Zhihu (知乎) — Chinese Quora (200 million+ registered users). Since 2018, I have been officially recognized as a “Top Zhihu Writer” (优秀回答者) in four fields (AI, ML, DM, and STAT). My articles have been read by more than 20,000,000 times. See my Zhihu page (微调).
Ph.D. Student in Information Systems and Management, 2019-2024
Carnegie Mellon University
M.S. in Applied Computing, 2015-2017
University of Toronto
B.S. in Computer Engineering (Minor in Computer Science and Math), 2015
University of Cincinnati
High School Diploma, 2010
Shanxi Experimental Secondary School 山西省实验中学
Jun 2021: Two impactful large-scale ML initiatives are under submission at NeurIPS 2021 (Datasets and Benchmarks). Please check out and follow them on OpenReview: (1) Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development and (2) Revisiting Time Series Outlier Detection: Definitions and Benchmarks.
Apr 2021: How to evaluate/select outlier detection models without any external information (e.g., ground truth)? We have a new preprint on using internal strategies for model selection. Do they suffice? Check out our paper!
I am open to peer review and organizing chances in the field of outlier & anomaly detection, ensemble Learning, clustering, ML libraries & systems, and information systems.
[w21f] Revisiting Time Series Outlier Detection: Definitions and Benchmarks, with Kwei-Herng Lai, Daochen Zha, Junjie Xu, Guanchu Wang, Xia Hu. Submitted to a major CS conference, under review. Preprint.
[w21e] Automatic Unsupervised Outlier Model Selection, with Ryan A. Rossi and Leman Akoglu. Submitted to a major CS conference, under review. Preprint.
[w21d] Copula-Based Outlier Detection, with Zheng Li, Xiyang Hu, Nicola Botta, Cezar Ionescu, and George H. Chen. Submitted to a key ML journal, under review. Preprint.
[w21c] A Large-scale Study on Unsupervised Outlier Model Selection: Do Internal Strategies Suffice? with Martin Q. Ma (equal contribution), Xiaorong Zhang, and Leman Akoglu. Preprint.
[w21b] Therapeutics Data Commons: Machine Learning Datasets and Tasks for Therapeutics, with Kexin Huang, Tianfan Fu, Wenhao Gao, Yusuf Roohani, Jure Leskovec, Connor W. Coley, Cao Xiao, Jimeng Sun, Marinka Zitnik. Submitted to a major CS conference, under review. Preprint.
[w21a] PyHealth: A Python Library for Health Predictive Models, with Zhi Qiao (equal contribution), Cao (Danica) Xiao, Lucas M. Glass, and Jimeng Sun. Preprint.
Designed new machine learning systems and models in healthcare.
Supervised by Dr. Cao (Danica) Xiao (IQVIA) and Prof. Jimeng Sun (UIUC).
Applied research in people analytics with machine learning.
Supervised by Prof. Anthony Bonner and the project is partly supported by Mitacs-Accelerate Research and Development Funding (IT07884).
I am happy to give talks on the series of tools I built, e.g., PyOD, combo, and SUOD. I am also willing to share my experience as a ML developer and researcher, especially on how to build ML tools from design. Please drop me a line for invite :)
I am an enthusiastic open-source developer: I build machine learning libraries and systems. Specifically, I initialized Python Outlier Detection library (PyOD) in 2018, which has become the most popular Python outlier detection toolkit. I also initialized combo: A Python Toolbox for Machine Learning Model Combination in July 2019–it is currently under active development.
I am currently working on a new ML system called SUOD (Scalable Unsupervised Outlier Detection), for accelerating model training and prediction when a large number of outlier detectors are presented on large, high-dimensional datasets. Watch/Star/Follow welcome!