avatar

Wenjie Xi

Ph.D. Candidate
George Mason University
wxi@gmu.edu


About Me

I am a Ph.D. candidate in Computer Science at Volgenau School of Engineering, George Mason University, advised by Professor Jessica Lin. I also spent a wonderful time as a Machine Learning Intern at Baidu in 2021, contributing to a computer vision project. I received my B.E. degree in Mechanical Engineering from Shanghai Normal University and M.S. degree in Computer Information Systems from Boston University.

Please email me for the latest CV/resume.

Research Interests

My research interests lie in machine learning, deep learning, and time series data mining. I focus on developing efficient and accurate deep learning models for time series classification, clustering, forecasting, and anomaly detection, in both univariate and multivariate settings. I am also exploring LLM for time series, as well as video anomaly detection and AI agents. My research also includes interdisciplinary applications of time series such as astrophysics, healthcare and cybersecurity, and interpretable machine learning.

News

Publications

  1. ECML-PKDD'25
    RandomAD: A Random Kernel-based Anomaly Detector for Time Series
    Wenjie Xi, Jessica Lin
    European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2025
    PDF Oral Presentation

  2. Preview
    VSFormer: Value and Shape-Aware Transformer with Prior-Enhanced Self-Attention for Multivariate Time Series Classification
    Wenjie Xi, Rundong Zuo, Alejandro Alvarez, Jie Zhang, Byron Choi, Jessica Lin
    Preview
    PDF

  3. DMKD
    RandomNet: Clustering Time Series Using Untrained Deep Neural Networks
    Xiaosheng Li*, Wenjie Xi*, Jessica Lin (*Co-first author)
    Data Mining and Knowledge Discovery (DMKD)
    PDF Code BibTex

  4. PAKDD
    Efficient and Accurate Similarity-Aware Graph Neural Network for Semi-supervised Time Series Classification
    Wenjie Xi, Arnav Jain, Li Zhang, Jessica Lin
    Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2024.
    PDF Code BibTex Oral Presentation

Services

Conference PC Members and Reviewers

Journal Reviewers

Teaching and mentoring


Powered by Jekyll and Minimal Light theme.