Bayesian methods deal with explicit assumptions and provide rules for reasoning consistently given those assumptions. Bayesian inferences are subjective in the sense that it is not plausible to reason about data without making assumptions. Bayesian NN learning from data features (i) background information used to select a prior probability distribution for the model parameters, and (ii) predictions of future observations performed by integrating the model's predictions with respect to the posterior parameter distribution obtained by updating this prior on the base of the newly acquired data.