What is it about?
Transformers in text are not directly applicable to human mobility modeling tasks, such as generating synthetic trajectories and next visit prediction, because of various difference between text and human mobility data. We utilize probablistic modeling to capture common sense and to enable Transformers for human trajectory generation.
Featured Image
Read the Original
This page is a summary of: TrajGPT: Controlled Synthetic Trajectory Generation Using a Multitask Transformer-Based Spatiotemporal Model, October 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3678717.3691303.
You can read the full text:
Contributors
The following have contributed to this page
