Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Bayesian Perceptron: Towards fully Bayesian Neural Networks

: Huber, Marco


Institute of Electrical and Electronics Engineers -IEEE-; IEEE Control Systems Society; Society for Industrial and Applied Mathematics -SIAM-, Philadelphia/Pa.:
59th IEEE Conference on Decision and Control, CDC 2020 : December 14th-18th 2020, Virtuell, Jeju Island, Korea (South)
Piscataway, NJ: IEEE, 2020
ISBN: 978-1-7281-7448-8
ISBN: 978-1-7281-7446-4
ISBN: 978-1-7281-7447-1
Conference on Decision and Control (CDC) <59, 2020, Online>
Fraunhofer IPA ()
Bayesian network; Künstliche Intelligenz; maschinelles Lernen; neuronales Netz; Unsicherheit

Artificial neural networks (NNs) have become the de facto standard in machine learning. They allow learning highly nonlinear transformations in a plethora of applications. However, NNs usually only provide point estimates without systematically quantifying corresponding uncertainties. In this paper a novel approach towards fully Bayesian NNs is proposed, where training and predictions of a perceptron are performed within the Bayesian inference framework in closed-form. The weights and the predictions of the perceptron are considered Gaussian random variables. Analytical expressions for predicting the perceptron’s output and for learning the weights are provided for commonly used activation functions like sigmoid or ReLU. This approach requires no computationally expensive gradient calculations and further allows sequential learning.