Pengyu Wang(王鹏宇)

I am currently pursuing the M.S. degree in software engineering at University of Electronic Science and Technology of China (UESTC) supervised by Prof. Fan Zhou.

My recent research interests include self-supervised learning, spatio-temporal data mining, deep generative model, urban computing.

Email  /  Google Scholar  /  Github /  CV

profile photo
Education

University of Electronic Science and Technology of China (UESTC), China
Bachelor Degree in School of Information and Software Engineering      • Sep. 2016 - Jun. 2020

University of Electronic Science and Technology of China (UESTC), China
Third-year Master, School of Information and Software Engineering      • Sep. 2020 - Present
Supervisor: Prof. Fan Zhou.

Publications
Contrastive Trajectory Learning for Tour Recommendation
Fan Zhou, Pengyu Wang, Xovee Xu, Wenxin Tai, Goce Trajcevski
ACM Transactions on Intelligent Systems and Technology, 2021

We presents a novel tour recommendation model by distilling knowledge and supervision signals from the trips in a self-supervised manner.

Improving Session-based Recommendation with Contrastive Learning
Wenxin Tai, Tian Lan, Zufeng Wu, Pengyu Wang, Yixiang Wang, Fan Zhou
User Modeling and User-Adapted Interaction, 2022

We propose a novel self-supervised session-based recommendation model that explores the additional signals using contrastive learning.

Self-supervised Human Mobility Learning for Next Location Prediction and Trajectory Classification
Fan Zhou, Yurou Dai, Qiang Gao, Pengyu Wang, Ting Zhong
Knowledge-Based Systems, 2021

We present a self-supervised mobility leaning framework to encode human mobility semantics and facilitate the downstream location-based tasks.

Diffusion Probabilistic Modeling for Fine-Grained Urban Traffic Flow Inference with Relaxed Structural Constraint
Xovee Xu, Yutao Wei Pengyu Wang, Xucheng Luo, Fan Zhou, Goce Trajcevski
ICASSP, 2023

We propose a novel probabilistic model DP-TFI to handle the urban flow uncertainties in fine-grained urban traffic flow inference problem and relax the structural constraint in a disentangled manner.

© 2023 Pengyu Wang | Powered by Jon Barron's design | Updated 2023-02-26