Hi there, I'm Yue ZHAO (赵越 in Chinese)! 👋

🌱 Short Bio: My name is Yue ZHAO (赵越 in Chinese). I am a third-year Ph.D. student at Heinz College, Carnegie Mellon University (CMU). Before joining CMU, I earned my Master degree from University of Toronto, and worked as a senior consultant at PwC Canada. I have coauthored more than 20 papers on anomaly detection, automated ML, and machine learning systems in leading venues. Service-wide, I am on the conference program committee of KDD, AAAI, and IJCAI, and reviewing for JMLR, TPAMI, and TKDE. I am a recipient of the 2022 Norton Labs Graduate Fellowship (one of the two graduate students worldwide). See my homepage and CV for more information.

I consider myself as an expert in outlier detection systems (ODSys)---I build automated, scalable ODSys to support real-world applications in security and finance with millions of downloads. I designed CPU-based (PyOD), GPU-based (TOD), distributed detection systems (SUOD) for tabular, time-series (TODS), and graph data (PyGOD).

Community: I am organizing the largest open-source community of outlier detection, including (1) the most popular detection system PyOD (6M downloads, top 10 data mining projects on GitHub) (2) the most watched and starred Anomaly Detection Resources with hundreds of books, tutorial, and papers (5k+ stars) (3) the discussion groups with hundreds of outlier detection researchers and practitioners from MIT, Meta, and more (see contact section below for joining).

🔭 Research-wise, I specialize in designing and building automated, scalable, and accelerated machine learning systems (MLSys), with realization and applications in outlier detection, healthcare, graph neural networks, and ensemble learning.

• Outlier detection systems (JMLR'19, AAAI'21, MLsys'21, NeurIPS'21)
• Outlier detection algorithms (IJCNN'18, SDM'19, BigData'20, ICDM'20, TKDE'22)
• Automated outlier detection (NeuIPS'21)
• AI x Science (AAAI'20, NeuIPS'21)

At CMU, I work with Prof. Leman Akoglu from DATA Lab on outlier detection, Prof. Zhihao Jia from Catalyst on machine learning systems, and Prof. George H. Chen on general ML and statistic methods. Externally, I am also fortunate to visit and collaborate with Prof. Jure Leskovec at Stanford University.

⚡ Open-source Contribution: I have led or contributed as a core member to more than 10 ML open-source initiatives, receiving 12,000 GitHub stars (top 0.002%: ranked 900 out of 40M GitHub users) and >500,0000 total downloads. Popular ones:

• [JMLR] PyOD: A Python Toolbox for Scalable Outlier Detection (Anomaly Detection).
• TOD: Tensor-based outlier detection--First large-scale GPU-based system for acceleration!
• [MLSys] SUOD: An Acceleration System for Large-scale Heterogeneous Outlier Detection.
• anomaly-detection-resources: The most starred resources (books, courses, etc.)!
• PyTorch Geometric (PyG): Graph Neural Network Library for PyTorch. Contributed to profiler & benchmarking, and heterogeneous data transformation, as a member of the PyG team.
• Python Graph Outlier Detection (PyGOD): A Python Library for Graph Outlier Detection.
• [NeurIPS] Therapeutics Data Commons (TDC): Machine learning for drug discovery.
• [AAAI] combo: A Python Toolbox for ML Model Combination (Ensemble Learning).
• [NeurIPS, AAAI] TODS: Time-series Outlier Detection. Contributed to core detection models.
• [NeurIPS] MetaOD: Automatic Unsupervised Outlier Model Selection (AutoML).

😄 I am open to:

• collaboration opportunities (anytime & anywhere & any type)
• paper review, tutorial, workshop, and talk opportunities

📫 Contact me by:

• Email (zhaoy [AT] cmu.edu)
• Twitter
• 知乎:「微调」
• 微信 @ 阿调
• Homepage

💬 News & Travel:

• Apr 2022: PyGOD (Python Graph Outlier Detection) received 300+ stars in a week!

• Apr 2022: We released PyGOD (Python Graph Outlier Detection). With PyGOD, you could do anomaly detection with the latest graph neural networks in 5 lines!

• Mar 2022: Invited to present at Morgan Stanley for large-scale anomaly detection systems!

• Mar 2022: 🎉 I received the prestigious 2022 Norton Labs Graduate Fellowship (one of the two graduate students worldwide). Thanks to the selection committee and my advisors!

• Mar 2022: ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions is accepted to IEEE Transactions on Knowledge and Data Engineering (TKDE)! ECOD is a simple yet effective detection algorithm with extremely fast O(nd) runtime.

• Feb 2022: 🌟 Reached 700 citations on Google Scholar!

• Feb 2022: Propose a new initiative called Detected AI (detected.ai) for large-scale anomaly detection applications. It is still too early to tell, but it will be exciting!

• Feb 2021: Have new system out TOD: GPU-accelerated Outlier Detection via Tensor Operations*. with George H. Chen and Zhihao Jia. Preprint, Code being released

  • TOD is the first fast, comprehensive, GPU-based outlier detection system.
  • 🌟 on average it is 11 times faster than PyOD!
  • 🌟 it supports various OD algorithms, e,g., kNN, LOF, ABOD, HBOS, etc.