Mass vaccination is one of the most effective epidemic control measures. Because one's vaccination decision is shaped by social processes, the pattern of vaccine uptake tends to show strong social and spatial heterogeneity, such as urban-rural divide and clustering. Examining through network perspectives, we develop a framework for quantifying the impact of spatial vaccination heterogeneity on epidemic outbreaks.
Leveraging fine-grained mobility data and computational models, we investigate two network effects---the ``hub effect'' (vaccinating mobility hubs reduces transmission) and the ``homophily effect'' (stronger homophily in vaccination rates increases transmission). Applying Bayesian deep learning and fine-grained epidemic simulations, we show a negative effect of homophily and a positive effect of highly vaccinated hubs on reducing case counts for both the synthetic network and the U.S. mobility network. Our framework enables us to evaluate the outcome of various hypothetical spatial vaccine distributions and propose a vaccination campaign strategy that targets a small number of regions with the largest gain in protective power, given the state of the COVID-19 as of January 2022. Our simulation shows that our strategy can prevent about 2.5 times more cases than a uniform strategy with an additional 1\% of the population vaccinated. Our study suggests that we must examine the interplay between vaccination patterns and mobility networks beyond the overall vaccination rate, and that accurate location-based targeting can be just as important as improving the overall vaccination rate.