Generating Individual Trajectories Using GPT-2 Trained from Scratch on Encoded Spatiotemporal Data
This work addresses trajectory generation for urban planning or mobility analysis, but it is incremental as it builds directly on prior research with a similar tokenization approach.
The paper tackles the problem of generating realistic individual daily trajectories by encoding spatiotemporal data as tokens and training a GPT-2 model from scratch, achieving a model that can generate trajectories influenced by environmental factors and individual attributes.
Following Mizuno, Fujimoto, and Ishikawa's research (Front. Phys. 2022), we transpose geographical coordinates expressed in latitude and longitude into distinctive location tokens that embody positions across varied spatial scales. We encapsulate an individual daily trajectory as a sequence of tokens by adding unique time interval tokens to the location tokens. Using the architecture of an autoregressive language model, GPT-2, this sequence of tokens is trained from scratch, allowing us to construct a deep learning model that sequentially generates an individual daily trajectory. Environmental factors such as meteorological conditions and individual attributes such as gender and age are symbolized by unique special tokens, and by training these tokens and trajectories on the GPT-2 architecture, we can generate trajectories that are influenced by both environmental factors and individual attributes.