arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2411.14768cs.LGcs.AI
收藏

网格与道路表达对于轨迹表示学习具有互补性

Grid and Road Expressions Are Complementary for Trajectory Representation Learning

University of Electronic Science and Technology of China(电子科技大学);Aalborg University(奥尔堡大学)

Silin Zhou, Shuo Shang, Lisi Chen, Peng Han, Christian S. Jensen

AI总结:

本文提出多模态轨迹表示学习方法GREEN,联合网格与道路轨迹表达,通过对比学习、掩码重建和交叉注意力融合互补信息,在多个下游任务上显著优于现有方法。

摘要:

轨迹表示学习(TRL)将轨迹映射为可用于许多下游任务的向量。现有TRL方法要么使用网格轨迹(捕捉自由空间中的移动),要么使用道路轨迹(捕捉道路网络中的移动)作为输入。我们观察到这两类轨迹具有互补性:一类提供区域和位置信息,另一类提供道路结构和移动规律。因此,我们提出一种新颖的多模态TRL方法,称为GREEN,以联合利用网格和道路轨迹表达进行有效的表示学习。具体来说,我们将原始GPS轨迹同时转换为网格轨迹和道路轨迹,并定制两个编码器以分别捕捉它们的信息。为了使两个编码器相互对齐并互补,我们采用对比损失促使它们为同一原始轨迹生成相似的嵌入,并设计掩码语言模型(MLM)损失,利用网格轨迹帮助重建被掩码的道路轨迹。为学习最终的轨迹表示,使用双模态交互器通过交叉注意力融合两个编码器的输出。我们将GREEN与7种最先进的TRL方法在3个下游任务上进行比较,发现GREEN一致优于所有基线,并将表现最佳基线的准确率平均提高了15.99%。

英文摘要:

Trajectory representation learning (TRL) maps trajectories to vectors that can be used for many downstream tasks. Existing TRL methods use either grid trajectories, capturing movement in free space, or road trajectories, capturing movement in a road network, as input. We observe that the two types of trajectories are complementary, providing either region and location information or providing road structure and movement regularity. Therefore, we propose a novel multimodal TRL method, dubbed GREEN, to jointly utilize Grid and Road trajectory Expressions for Effective representatioN learning. In particular, we transform raw GPS trajectories into both grid and road trajectories and tailor two encoders to capture their respective information. To align the two encoders such that they complement each other, we adopt a contrastive loss to encourage them to produce similar embeddings for the same raw trajectory and design a mask language model (MLM) loss to use grid trajectories to help reconstruct masked road trajectories. To learn the final trajectory representation, a dual-modal interactor is used to fuse the outputs of the two encoders via cross-attention. We compare GREEN with 7 state-of-the-art TRL methods for 3 downstream tasks, finding that GREEN consistently outperforms all baselines and improves the accuracy of the best-performing baseline by an average of 15.99\%.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑