top of page

✦✦UPCOMING✦✦ 12 JUN 2026 (FRI) 10:00-12:00

  • 2 days ago
  • 2 min read

Geography Distinguished Seminar Series

Research Frontier of Trajectory Similarity Learning


Date: 12 JUN 2026 (Friday)

Time: 10:00-12:00 (HKT)

Venue: Chamber, Faculty of Social Sciences, 11/F, The Jockey Club Tower, Centennial Campus, HKU

Mode: Hybrid

Via Zoom: Zoom link will be provided upon successful registration


Abstract:

Trajectory similarity computation is a fundamental functionality that is used for clustering, prediction, anomaly detection, and many other applications. Traditional trajectory similarity measurements involve point-to-point distance computation for two trajectories and are computationally expensive. In contrast, trajectory similarity learning (TSL) models represent variable-length trajectories as fixed-size representation vectors, for which the similarity computation is much faster. This talk will briefly review the state-of-the-art of TSL and then report on our recent studies in this research area. Our studies generally focus on TSL efficiency and effectiveness across both unsupervised and supervised learning settings. Our main technical contributions include adaptive contrastive loss function design, movement semantics modeling, road network representation, and efficiency-oriented architecture designs. Our proposals outperform state-of-the-art in terms of multiple performance metrics including effectiveness and efficiency. The talk will end with discussion on possible future research directions for TSL.

Professor Hua Lu

Professor, Department of Computer Science, University of Aalborg, Denmark

Professor Hua Lu is a Professor at the Department of Computer Science, Aalborg University, Denmark. He works in the university’s Copenhagen campus, where he leads research on Data Engineering, Science and Systems. He received the BSc and MSc degrees from Peking University, China, and the PhD degree in computer science from National University of Singapore. His research spans data management, spatial data and GeoAI, location-based services, and data science. He has served as PC cochair or vice chair for ISA 2011, MUE 2011, MDM 2012, NDBC 2019, IEEE BigData 2022, SSTD 2025, and PAKDD 2026, demo chair for SSDBM 2014, PhD forum cochair for MDM 2016 and MDM 2022, and senior PC member for CIKM 2024, WWW 2025, and ICDE 2027. He has also served on the program committees for conferences such as SIGMOD, VLDB, ICDE, EDBT, SIGKDD, SIGIR, WWW, IJCAI, CIKM, WSDM, ACM SIGSPATIAL, SSTD and MDM. He received among others the Best Vision Paper Award at SSTD 2019, Best Paper Runner-Up Award at IEEE MDM 2024, Distinguished Reviewer Award at IEEE ICDE 2026, and Outstanding Reviewer Award at IEEE ICDE 2024. He is a senior member of the IEEE.

HKU Geography seminar poster with Prof. Hua Lu, neural network graphic, QR code, and details for 12 Jun 2026, 10:00–12:00.

 
 
Textured Paper Surface

 INTERDISCIPLINARY INITIATIVES 

GGAS.png
GEOG1931 copy_edited_edited.png
cube logo.png
glass logo -black.png
ICCDS_edited.png
ITS (new).png
JLFC [logo] -1.png
  • Instagram
  • WeChat
  • Facebook
  • Youtube
  • x

HKU Geography

Home

General Enquiry: 
Room 10.23
geog@hku.hk

UG Enquiry & Classwork Submission:
Room 10.48
geogug@hku.hk

10F, The Jockey Club Tower,
Centennial Campus,
Pokfulam Road, Hong Kong

CoWIN logo.png

Subscribe to Our Newsletter

© 2026 by Department of Geography, The University of Hong Kong.

bottom of page