This paper presents a probabilistic framework to obtain both reliable and fast uncertainty estimates for predictions with Deep Neural Networks (DNNs). Our main contribution is a practical and principled combination of DNNs with sparse Gaussian Processes (GPs). We prove theoretically that DNNs can be seen as a special case of sparse GPs, namely mixtures of GP experts (MoE-GP), and we devise a learning algorithm that brings the derived theory into practice. In experiments from two different robotic tasks – inverse dynamics of a manipulator and object detection on a micro-aerial vehicle (MAV) – we show the effectiveness of our approach in terms of predictive uncertainty, proved scalability, and runtime efficiency on a Jetson TX2. We thus argue that our approach can pave the way towards reliable and fast robot learning systems with uncertainty wareness.
Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes
2021-11-08
Conference paper
Electronic Resource
English
Estimating Model Uncertainty of Neural Networks in Sparse Information Form
German Aerospace Center (DLR) | 2020
|SAE Technical Papers | 2021
|Laplace Approximation for Uncertainty Estimation of Deep Neural Networks
German Aerospace Center (DLR) | 2019
|