Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

GATGPT: A Pre-trained Large Language Model with Graph Attention Network for Spatiotemporal Imputation

About

The analysis of spatiotemporal data is increasingly utilized across diverse domains, including transportation, healthcare, and meteorology. In real-world settings, such data often contain missing elements due to issues like sensor malfunctions and data transmission errors. The objective of spatiotemporal imputation is to estimate these missing values by understanding the inherent spatial and temporal relationships in the observed multivariate time series. Traditionally, spatiotemporal imputation has relied on specific, intricate architectures designed for this purpose, which suffer from limited applicability and high computational complexity. In contrast, our approach integrates pre-trained large language models (LLMs) into spatiotemporal imputation, introducing a groundbreaking framework, GATGPT. This framework merges a graph attention mechanism with LLMs. We maintain most of the LLM parameters unchanged to leverage existing knowledge for learning temporal patterns, while fine-tuning the upper layers tailored to various applications. The graph attention component enhances the LLM's ability to understand spatial relationships. Through tests on three distinct real-world datasets, our innovative approach demonstrates comparable results to established deep learning benchmarks.

Yakun Chen, Xianzhi Wang, Guandong Xu• 2023

Related benchmarks

TaskDatasetResultRank
Spatiotemporal Traffic ForecastingMilan-Internet
NRMSE0.1323
63
Spatiotemporal Traffic ForecastingTrentino INTERNET
NRMSE0.6354
24
Spatiotemporal Traffic ForecastingMilan-SMS
NRMSE0.8515
24
Spatiotemporal Traffic ForecastingTrentino-SMS
NRMSE1.4071
24
Traffic PredictionTrentino SMS IN
MAE0.7211
24
Traffic PredictionTrentino INTERNET
MAE1.2963
20
Spatio-temporal ImputationTrentino INTERNET
MAE0.8166
13
Spatio-temporal ImputationTrentino SMS-OUT
MAE0.6243
13
Spatio-temporal traffic forecastingMilan-Internet
MAE89.1961
13
ImputationMilan telecommunications SMS-IN
MAE0.1289
13
Showing 10 of 31 rows

Other info

Follow for update