Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Yuyao, Yang, Min, Chen, Meng, Huang, Weiming, Yin, Yilong, Gong, Yongshun
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.07383
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911663814344704
author Wang, Yuyao
Yang, Min
Chen, Meng
Huang, Weiming
Yin, Yilong
Gong, Yongshun
author_facet Wang, Yuyao
Yang, Min
Chen, Meng
Huang, Weiming
Yin, Yilong
Gong, Yongshun
contents Cross-city transfer improves prediction in label-scarce cities by leveraging labeled data from other cities, but it becomes challenging when cities adopt incompatible partitions and no ground-truth region correspondences exist. Existing approaches either rely on heuristic region matching, which is often sensitive to anchor choices, or perform distribution-level alignment that leaves correspondences implicit and can be unstable under strong heterogeneity. We propose SCOT, a cross-city representation learning framework that learns explicit soft correspondences between unequal region sets via Sinkhorn-based entropic optimal transport. SCOT further sharpens transferable structure with an OT-weighted contrastive objective and stabilizes optimization through a cycle-style reconstruction regularizer. For multi-source transfer, SCOT aligns each source and the target to a shared prototype hub using balanced entropic transport guided by a target-induced prototype prior. Across real-world cities and tasks, SCOT consistently improves transfer accuracy and robustness, while the learned transport couplings and hub assignments provide interpretable diagnostics of alignment quality.
format Preprint
id arxiv_https___arxiv_org_abs_2604_07383
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SCOT: Multi-Source Cross-City Transfer with Optimal-Transport Soft-Correspondence Objective
Wang, Yuyao
Yang, Min
Chen, Meng
Huang, Weiming
Yin, Yilong
Gong, Yongshun
Machine Learning
Cross-city transfer improves prediction in label-scarce cities by leveraging labeled data from other cities, but it becomes challenging when cities adopt incompatible partitions and no ground-truth region correspondences exist. Existing approaches either rely on heuristic region matching, which is often sensitive to anchor choices, or perform distribution-level alignment that leaves correspondences implicit and can be unstable under strong heterogeneity. We propose SCOT, a cross-city representation learning framework that learns explicit soft correspondences between unequal region sets via Sinkhorn-based entropic optimal transport. SCOT further sharpens transferable structure with an OT-weighted contrastive objective and stabilizes optimization through a cycle-style reconstruction regularizer. For multi-source transfer, SCOT aligns each source and the target to a shared prototype hub using balanced entropic transport guided by a target-induced prototype prior. Across real-world cities and tasks, SCOT consistently improves transfer accuracy and robustness, while the learned transport couplings and hub assignments provide interpretable diagnostics of alignment quality.
title SCOT: Multi-Source Cross-City Transfer with Optimal-Transport Soft-Correspondence Objective
topic Machine Learning
url https://arxiv.org/abs/2604.07383