English

Spatio-Temporal Few-Shot Learning via Diffusive Neural Network Generation

Machine Learning 2024-03-26 v3

Abstract

Spatio-temporal modeling is foundational for smart city applications, yet it is often hindered by data scarcity in many cities and regions. To bridge this gap, we propose a novel generative pre-training framework, GPD, for spatio-temporal few-shot learning with urban knowledge transfer. Unlike conventional approaches that heavily rely on common feature extraction or intricate few-shot learning designs, our solution takes a novel approach by performing generative pre-training on a collection of neural network parameters optimized with data from source cities. We recast spatio-temporal few-shot learning as pre-training a generative diffusion model, which generates tailored neural networks guided by prompts, allowing for adaptability to diverse data distributions and city-specific characteristics. GPD employs a Transformer-based denoising diffusion model, which is model-agnostic to integrate with powerful spatio-temporal neural networks. By addressing challenges arising from data gaps and the complexity of generalizing knowledge across cities, our framework consistently outperforms state-of-the-art baselines on multiple real-world datasets for tasks such as traffic speed prediction and crowd flow prediction. The implementation of our approach is available: https://github.com/tsinghua-fib-lab/GPD.

Keywords

Cite

@article{arxiv.2402.11922,
  title  = {Spatio-Temporal Few-Shot Learning via Diffusive Neural Network Generation},
  author = {Yuan Yuan and Chenyang Shao and Jingtao Ding and Depeng Jin and Yong Li},
  journal= {arXiv preprint arXiv:2402.11922},
  year   = {2024}
}
R2 v1 2026-06-28T14:52:49.124Z