GTG: Generalizable Trajectory Generation Model for Urban Mobility

Authors: Jingyuan Wang, Yujing Lin, Yudong Li

AAAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments on three datasets demonstrates that our model significantly outperforms existing models in terms of generalization ability.
Researcher Affiliation Academia 1School of Computer Science and Engineering, Beihang University, Beijing, China 2MIIT Key Laboratory of Data Intelligence and Management, Beihang University, Beijing, China 3School of Economics and Management, Beihang University, Beijing, China
Pseudocode No The paper describes its methodology using mathematical formulations and descriptive text, but it does not include any explicitly labeled pseudocode blocks or algorithms with structured, step-by-step procedures in a code-like format.
Open Source Code Yes Code https://github.com/lyd1881310/GTG
Open Datasets No Three real-world trajectory datasets are used to evaluate the performance of our proposed method. These datasets were collected in the three cities, namely Beijing(BJ), Xi an(XA) and Chengdu(CD). Road network of the three cities are collected from the Open Street Map (Open Street Map contributors 2017), and road segment trajectories are obtained by performing mapmatching algorithm (Yang and Gidofalvi 2018).
Dataset Splits No The paper describes the datasets used (Beijing, Xi'an, Chengdu) and mentions preprocessing steps like map-matching, but it does not specify any training, validation, or test dataset splits (e.g., percentages or exact counts).
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., CPU, GPU models, memory) used to run the experiments.
Software Dependencies No The paper mentions various methods and frameworks such as Inductive GCN (Hamilton, Ying, and Leskovec 2017), GATv2 (Brody, Alon, and Yahav 2022), METIS (Chiang et al. 2019), and Lib City (Wang et al. 2021a), but it does not provide specific version numbers for these or other ancillary software components required for replication.
Experiment Setup No The paper describes the model architecture and loss functions, including balance weights like λr and λg. However, it does not provide specific hyperparameter values (e.g., learning rates, batch sizes, number of epochs) or other detailed training configurations used for the experiments.