Spatio-Temporal Contrastive Self-Supervised Learning for POI-level Crowd Flow Inference

no code implementations • 6 Sep 2023 • Songyu Ke, Ting Li, Li Song, Yanping Sun, Qintian Sun, Junbo Zhang, Yu Zheng

To address these challenges, we recast the crowd flow inference problem as a self-supervised attributed graph representation learning task and introduce a novel Contrastive Self-learning framework for Spatio-Temporal data (CSST).

Contrastive Learning Graph Representation Learning +2

AirFormer: Predicting Nationwide Air Quality in China with Transformers

1 code implementation • 29 Nov 2022 • Yuxuan Liang, Yutong Xia, Songyu Ke, Yiwei Wang, Qingsong Wen, Junbo Zhang, Yu Zheng, Roger Zimmermann

Air pollution is a crucial issue affecting human health and livelihoods, as well as one of the barriers to economic and social growth.