Classified as unknown: A novel Bayesian neural network

01/31/2023

∙

by   Tianbo Yang, et al.

∙

0

∙

share

We establish estimations for the parameters of the output distribution for the softmax activation function using the probit function. As an application, we develop a new efficient Bayesian learning algorithm for fully connected neural networks, where training and predictions are performed within the Bayesian inference framework in closed-form. This approach allows sequential learning and requires no computationally expensive gradient calculation and Monte Carlo sampling. Our work generalizes the Bayesian algorithm for a single perceptron for binary classification in <cit.> to multi-layer perceptrons for multi-class classification.