Satellite observations are widely used to estimate the concentrations of surface air pollutants, but the temporal coverage of these datasets is relatively short. To overcome this limitation, we propose a wide–deep ensemble machine learning framework to reconstruct the fine particulate matter (particulate matter lower than 2.5 μm (PM_2.5)) dataset of East Asia (EA) over the past four decades (1981–2020). The results indicate that the framework effectively leveraged the advantages of satellite observations (higher accuracy) and model-based estimations (longer temporal coverage) of surface air pollutants. The reconstructed PM_2.5 concentrations agreed well with the ground measurements, with coefficient of determination (R²) and root-mean-square error (RMSE) values of 0.99 and 1.38 μg·m⁻³, respectively, which outperformed the satellite-based PM_2.5 estimates. As more ground measurements were incorporated into the model for training, the average RMSE in Japan and the Korean Peninsula decreased to 0.83 and 1.50 μg·m⁻³, respectively. Simultaneously, on the basis of the reconstructed datasets, we investigated the exposure level to PM_2.5 in EA from 1981 to 2020. Since 2000, the increase in anthropogenic emissions has substantially worsened the air quality in EA, and nearly 50% of the population resided in areas where the annual average PM_2.5 concentrations exceeded 50 μg·m⁻³ from 2009 to 2010. Despite the implementation of various mitigation strategies by local authorities to lower the ambient PM_2.5 concentrations, the entire exposure level in EA is still implausible to meet the World Health Organization (WHO) air quality guidelines. In addition, population aging and climate change have the potential to increase PM_2.5 exposure risk in the future. For policy-makers in EA, it is essential to consider the effects of these factors and develop more effective mitigation strategies that aim to lessen the health impact associated with PM_2.5 exposure.