Balancing Imbalance: Data-Scarce Urban Flow Prediction via Spatio-Temporal Balanced Transfer Learning
Authors: Xinyan Hao, Huaiyu Wan, Shengnan Guo, Youfang Lin
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on real-world datasets validate the efficacy of our proposed approach compared with several state-of-the-art methods. |
| Researcher Affiliation | Academia | 1School of Computer Science and Technology, Beijing Jiaotong University, Beijing, China 2Beijing Key Laboratory of Traffic Data Mining and Embodied Intelligence, Beijing, China EMAIL. All authors are affiliated with Beijing Jiaotong University, which is an academic institution, and their email addresses use the .edu.cn domain, indicating an academic affiliation. |
| Pseudocode | No | The paper describes a "bidirectional complementary iterative learning algorithm" but does not present it in a structured pseudocode or algorithm block. The steps are described in paragraph text rather than a formatted algorithm. |
| Open Source Code | Yes | The code is made publicly available at https://github.com/Shaun Hao/stbat. |
| Open Datasets | Yes | We evaluate our proposed framework on real-world public datasets of three cities: New York (NY), Chicago (CHI), and Washington (DC), which contain vehicle pickup and drop-off records of bike and taxi. All data are collected and opened by [Jin et al., 2022]. Additionally, we also use public POI data of each city. |
| Dataset Splits | Yes | For each source city, we divided the training and validation sets in a 2:1 ratio. For each target city, we allocate the last 2 months for testing, the preceding 2 months for validation, and the last 3 days or 0 days before the validation data for training, forming two data-scarce scenarios: few-shot and zero-shot. |
| Hardware Specification | No | The paper mentions implementing STBaT using PyTorch but does not specify any particular CPU or GPU models, or other hardware details used for running the experiments. |
| Software Dependencies | No | We implement STBa T using Py Torch. The paper mentions PyTorch but does not provide a specific version number for PyTorch or any other software dependencies, such as Python or CUDA versions. |
| Experiment Setup | Yes | For the dimensions of FCs in the regional pattern autoencoder, we set them to (16, 32) for the encoder, (32, 14) for the POI classifier and (32, 256) for the spatio-temporal feature distribution predictor. We choose the ST-net model as the spatio-temporal feature extractor, stacking three residual blocks with 64 channels and a single-layer LSTM with 128 hidden units. For the STBLM, the dimensions of the FC layers of the feedforward predictor are set to 256 and 1, and the dimensions of the feedforward projector are set to 256 and 256. To capture temporal dependencies, we set the horizon ̕ to 6... We train the model until the validation error does not decrease for 5 consecutive epochs on the source data... (1) We vary the dimension Δof the region embedding, and set it to 32. (2) We tune the parameter Βof embedding loss and find that a value of 0.01 consistently yields the best performance. (3) We adjust the bandwidth Ηof the RIAM’s density estimator, and set it to 0.85 to achieve the best performance. (4) We set the temperature ̕dcl to 1 to better learn balanced spatio-temporal feature extraction. |