Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes
This addresses the need for uncertainty-aware robot learning systems, though it is incremental as it builds on existing methods like sparse GPs and DNNs.
The paper tackles the problem of obtaining reliable and fast uncertainty estimates for deep neural network predictions in robotics by combining DNNs with sparse Gaussian Processes, showing effectiveness in tasks like inverse dynamics and object detection with improved scalability and run-time efficiency on a Jetson TX2.
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, improved scalability, and run-time efficiency on a Jetson TX2. We thus argue that our approach can pave the way towards reliable and fast robot learning systems with uncertainty awareness.