elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes

Lee, Jongseok and Feng, Jianxiang and Humt, Matthias and Müller, Marcus Gerhard and Triebel, Rudolph (2021) Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes. In: 5th Conference on Robot Learning, CoRL 2021. Proceedings of Machine Learning Research (PMLR). 5th Conference on Robot Learning (CoRL), 2021-11-08 - 2021-11-11, London, United Kingdon. ISSN 2640-3498.

[img] PDF
2MB

Abstract

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.

Item URL in elib:https://elib.dlr.de/145805/
Document Type:Conference or Workshop Item (Poster)
Title:Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Lee, JongseokUNSPECIFIEDhttps://orcid.org/0000-0002-0960-0809UNSPECIFIED
Feng, JianxiangUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Humt, MatthiasUNSPECIFIEDhttps://orcid.org/0000-0002-1523-9335UNSPECIFIED
Müller, Marcus GerhardUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Triebel, RudolphUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Date:8 November 2021
Journal or Publication Title:5th Conference on Robot Learning, CoRL 2021
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:No
Editors:
EditorsEmailEditor's ORCID iDORCID Put Code
Neil, LawrenceUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Reid, MarkUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Publisher:Proceedings of Machine Learning Research (PMLR)
ISSN:2640-3498
Status:Published
Keywords:Robotic Introspection, Bayesian Deep Learning, Gaussian Processes
Event Title:5th Conference on Robot Learning (CoRL)
Event Location:London, United Kingdon
Event Type:international Conference
Event Start Date:8 November 2021
Event End Date:11 November 2021
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Robotics
DLR - Research area:Raumfahrt
DLR - Program:R RO - Robotics
DLR - Research theme (Project):R - Explainable Robotic AI, R - Intelligent Mobility (RM) [RO]
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Robotics and Mechatronics (since 2013)
Institute of Robotics and Mechatronics (since 2013) > Perception and Cognition
Deposited By: Lee, Jongseok
Deposited On:19 Nov 2021 08:58
Last Modified:19 Jul 2024 09:30

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.