Distributed multi-stage image compression—where visual content traverses multiple processing nodes under varying quality requirements—poses challenges. Progressive methods enable bitstream truncation but underutilize available compute resources; successive compression repeats costly pixel-domain operations and suffers cumulative quality loss and inefficiency; fixed-parameter models lack post-encoding flexibility. In this work, we developed the Hierarchical Cascade Framework (HCF) that achieves high rate-distortion performance and better computational efficiency through direct latent-space transformations across network nodes in distributed multi-stage image compression systems. Under HCF, we introduced policy-driven quantization control to optimize rate–distortion trade-offs, and established the edge quantization principle through differential entropy analysis. The configuration based on this principle demonstrates up to 0.6dB PSNR gains over other configurations. When comprehensively evaluated on the Kodak, CLIC, and CLIC2020-mobile datasets, HCF outperforms successive-compression methods by up to 5.56% BD-Rate in PSNR on CLIC, while saving up to 97.8% FLOPs, 96.5% GPU memory, and 90.0% execution time. It also outperforms state-of-the-art progressive compression methods by up to 12.64% BD-Rate on Kodak and enables retraining-free cross-quality adaptation with 7.13-10.87% BD-Rate reductions on CLIC2020-mobile.

In this work, we develop an agent-based model to predict the infection spread with accuracy. Under COVID-19 pandemic, we faced difficult decision-making of disease control including social distancing and shutdown, which significantly restricted our daily living. Without strong evidences to its effect, however, it was very hard to implement necessary policies in a timely manner. To this end, it is imperative to design a computationally efficient simulation model that can predict the infection spread with accuracy, taking into account the changes of important control policies. We develop an agent-based model that can incorporate individual behaviors and interactions, while capturing large-scale features of the infection spread. We verify the accuracy of our model by comparing the prediction results with the traces of COVID-19 confirmed cases in several countries.