Jul 09, 2020. Neural networks can be large, generally brimming with billions of parameters. technology information So it can be a heavy computational lift just to get an answer, let alone a confidence level. Uncertainty analysis in neural networks isn’t new. However previous approaches, stemming from Bayesian deep learning, have relied on working, or sampling, a neural community many instances over to understand its confidence. That course of takes time and reminiscence, a luxurious that may not exist in high-pace site visitors.