Accurate and trustworthy epidemic forecasting is an important problem that has impact on public health planning and disease mitigation. Most work such as deep sequential models have shown superior forecasting performance, but disregard uncertainty quantification resulting in mis-calibrated predictions. Existing work in neural models for time-series analysis also have several limitations; e.g. it is difficult to specify meaningful priors in Bayesian NN, while methods like deep ensembling are comp…