TU Darmstadt / ULB / TUprints

Online Learning with Stochastic Recurrent Neural Networks using Intrinsic Motivation Signals

Tanneberg, Daniel ; Peters, Jan ; Rueckert, Elmar (2022)
Online Learning with Stochastic Recurrent Neural Networks using Intrinsic Motivation Signals.
CoRL2017 - Conference on Robot Learning 2017. Mountain View, California (13.-15.11.2017)
doi: 10.26083/tuprints-00020580
Conference or Workshop Item, Secondary publication, Publisher's Version

[img] Text
corl17_01.pdf
Copyright Information: CC BY 4.0 International - Creative Commons, Attribution.

Download (1MB)
Item Type: Conference or Workshop Item
Type of entry: Secondary publication
Title: Online Learning with Stochastic Recurrent Neural Networks using Intrinsic Motivation Signals
Language: English
Date: 2022
Place of Publication: Darmstadt
Publisher: PMLR
Book Title: Proceedings of the 1st Annual Conference on Robot Learning
Series: Proceedings of Machine Learning Research
Series Volume: 78
Collation: 8 Seiten
Event Title: CoRL2017 - Conference on Robot Learning 2017
Event Location: Mountain View, California
Event Dates: 13.-15.11.2017
DOI: 10.26083/tuprints-00020580
Corresponding Links:
Origin: Secondary publication service
Abstract:

Continuous online adaptation is an essential ability for the vision of fully autonomous and lifelong-learning robots. Robots need to be able to adapt to changing environments and constraints while this adaption should be performed without interrupting the robot’s motion. In this paper, we introduce a framework for probabilistic online motion planning and learning based on a bio-inspired stochastic recurrent neural network. Furthermore, we show that the model can adapt online and sample-efficiently using intrinsic motivation signals and a mental replay strategy. This fast adaptation behavior allows the robot to learn from only a small number of physical interactions and is a promising feature for reusing the model in different environments. We evaluate the online planning with a realistic dynamic simulation of the KUKA LWR robotic arm. The efficient online adaptation is shown in simulation by learning an unknown workspace constraint using mental replay and cognitive dissonance as intrinsic motivation signal.

Uncontrolled Keywords: Lifelong-learning, Intrinsic Motivation, Recurrent Neural Networks
Status: Publisher's Version
URN: urn:nbn:de:tuda-tuprints-205803
Classification DDC: 000 Generalities, computers, information > 004 Computer science
Divisions: 20 Department of Computer Science > Intelligent Autonomous Systems
TU-Projects: EC/H2020|640554|SKILLS4ROBOTS
Date Deposited: 18 Nov 2022 14:30
Last Modified: 24 Mar 2023 09:29
URI: https://tuprints.ulb.tu-darmstadt.de/id/eprint/20580
PPN: 502453923
Export:
Actions (login required)
View Item View Item