About Me
Hello! I am Vincent, a PhD candidate in the Department of Civil Engineering at McGill University, working under the supervision of Prof. Lijun Sun since January 2021. My research focuses on probabilistic time series forecasting and machine learning. I hold a Bachelor’s and a Master’s degree in Transportation Engineering from Central South University (2017 and 2020, respectively), where I specialized in human mobility patterns and complex network analysis.
A Glimpse of My Work
Below is a selection of my key publications that I believe you will find valuable. For a full list of my research works, please visit my Google Scholar profile. I hope these papers offer insights into my research and contribute to your understanding of probabilistic time series forecasting and related fields.
- Enhancing Deep Traffic Forecasting Models with Dynamic Regression: This paper extends deep learning-based traffic forecasting by incorporating cross-correlated error modeling into a multivariate Seq2Seq framework using a matrix-variate autoregressive (AR) process. [arXiv]
- [NeurIPS 2024] Multivariate Probabilistic Time Series Forecasting with Correlated Errors: A follow-up to the “Better Batch” paper, this work extends the approach to multivariate time series forecasting by introducing an efficient and scalable parameterization of the cross-covariance matrix. [arXiv]
- [AISTATS 2024] Better Batch for Deep Probabilistic Time Series Forecasting: My first PhD project, where I introduced the GLS loss into the temporal domain to account for autocorrelated error terms in univariate probabilistic time series forecasting. The arXiv version includes corrections to the notation found in the PMLR version. [arXiv][PMLR]
- Understanding Coupling Dynamics of Public Transportation Networks: A key paper from my master’s research, offering an introduction to multiplex network theory in transportation. The methods section is particularly useful for guiding trip reconstruction using smart card data, though the region-specific results may vary in relevance for some readers. [Paper]
Milestones
- FRQNT B2X Scholarship, Fonds de recherche du Québec (FRQ), awarded C$77,000, 2022
- McGill Engineering Doctoral Award (MEDA), awarded C$111,000, 2021
- Accepted into PhD programs at University of Washington (Seattle) and McGill University, 2020
- Outstanding Graduate of Hunan Province (Master’s level), China, 2020
- National Scholarship for Graduate Students, China, 2018
- Outstanding Graduate of Hunan Province (Bachelor’s level), China, 2017
- National Scholarship for Outstanding Students in Transportation Engineering, China, 2016
- 1st Place Winner, National Competition of Transport Science and Technology, China, 2016
Teaching Experience
- TA, CIVE 440 Traffic Engineering and Simulation, Fall 2024, McGill University
- TA, CIVE 542 Transportation Network Analysis, Winter 2024, McGill University
- TA, CIVE 440 Traffic Engineering and Simulation, Fall 2023, McGill University
- TA, CIVE 542 Transportation Network Analysis, Winter 2023, McGill University
- TA, CIVE 319 Transportation Engineering, Winter 2022, McGill University
Work Experience
Project: Adaptive multi-horizon models for probabilistic demand forecasting
Sponsors
