Self-Supervised Cross-City Trajectory Representation Learning Based on Meta-Learning
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (19) · 16217-16225
DOI: 10.1609/aaai.v40i19.38658 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yanwei Yu und 5 weiteren Autor:innen.
Abstract (Originalsprache)
Trajectory representation learning transforms complex spatio-temporal features of trajectories into dense, low-dimensional embeddings, enabling applications in intelligent transportation systems. With advances in this field and the availability of large-scale traffic data, intelligent urban systems have been widely deployed in major cities. However, existing methods heavily rely on large volumes of trajectory data, limiting their transferability to cities with sparse data, especially small or less-developed ones. Moreover, most current approaches learn representations within a single city, overlooking the shared travel patterns across regions and cities with similar geographic contexts. To address these issues, we propose MetaTRL, a self-supervised cross-city trajectory representation learning method based on meta-learning. Specifically, we introduce a Shared and Private Parameterized Cross-city Meta-learning Framework to support knowledge sharing and transfer across cities. We further design a Meta-knowledge Enhanced Road Segment Encoder and a Trajectory Encoder that integrates private and shared knowledge to learn and fuse spatio-temporal trajectory features. Extensive experiments on two real-world datasets and multiple downstream tasks demonstrate the significant superiority of MetaTRL over state-of-the-art baselines and achieves a remarkable average improvement of 134.66% in Macro-F1 on destination prediction task.
Zitieren
APA
Yu, Y., Xia, H., Gu, S., Zhao, X., Chen, D., & Cao, Y. (2026). Self-Supervised Cross-City Trajectory Representation Learning Based on Meta-Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(19), 16217-16225. https://doi.org/10.1609/aaai.v40i19.38658
Vancouver
Yu Y, Xia H, Gu S, Zhao X, Chen D, Cao Y. Self-Supervised Cross-City Trajectory Representation Learning Based on Meta-Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(19):16217-16225. doi:10.1609/aaai.v40i19.38658
BibTeX
@article{Yu2026,
author = {Yu, Yanwei and Xia, Hong and Gu, Shaoxuan and Zhao, Xingyu and Chen, Dongliang and Cao, Yuan},
title = {Self-Supervised Cross-City Trajectory Representation Learning Based on Meta-Learning},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {19},
pages = {16217--16225},
doi = {10.1609/aaai.v40i19.38658}
}
EUVIMED