TY - JOUR

T1 - Intrinsic motivation and mental replay enable efficient online adaptation in stochastic recurrent networks

AU - Tanneberg, Daniel

AU - Peters, Jan

AU - Rueckert, Elmar

N1 - Publisher Copyright: © 2018 Elsevier Ltd

PY - 2019/1

Y1 - 2019/1

N2 - Autonomous robots need to interact with unknown, unstructured and changing environments, constantly facing novel challenges. Therefore, continuous online adaptation for lifelong-learning and the need of sample-efficient mechanisms to adapt to changes in the environment, the constraints, the tasks, or the robot itself are crucial. In this work, we propose a novel framework for probabilistic online motion planning with online adaptation based on a bio-inspired stochastic recurrent neural network. By using learning signals which mimic the intrinsic motivation signal cognitive dissonance in addition with a mental replay strategy to intensify experiences, the stochastic recurrent network can learn from few physical interactions and adapts to novel environments in seconds. We evaluate our online planning and adaptation framework on an anthropomorphic KUKA LWR arm. The rapid online adaptation is shown by learning unknown workspace constraints sample-efficiently from few physical interactions while following given way points.

AB - Autonomous robots need to interact with unknown, unstructured and changing environments, constantly facing novel challenges. Therefore, continuous online adaptation for lifelong-learning and the need of sample-efficient mechanisms to adapt to changes in the environment, the constraints, the tasks, or the robot itself are crucial. In this work, we propose a novel framework for probabilistic online motion planning with online adaptation based on a bio-inspired stochastic recurrent neural network. By using learning signals which mimic the intrinsic motivation signal cognitive dissonance in addition with a mental replay strategy to intensify experiences, the stochastic recurrent network can learn from few physical interactions and adapts to novel environments in seconds. We evaluate our online planning and adaptation framework on an anthropomorphic KUKA LWR arm. The rapid online adaptation is shown by learning unknown workspace constraints sample-efficiently from few physical interactions while following given way points.

KW - Autonomous robots

KW - Experience replay

KW - Intrinsic motivation

KW - Neural sampling

KW - Online learning

KW - Spiking recurrent networks

UR - http://www.scopus.com/inward/record.url?scp=85056187742&partnerID=8YFLogxK

UR - https://cps.unileoben.ac.at/wp/NeuralNetworks2018Tanneberg.pdf

U2 - 10.1016/j.neunet.2018.10.005

DO - 10.1016/j.neunet.2018.10.005

M3 - Article

C2 - 30408695

AN - SCOPUS:85056187742

VL - 109.2019

SP - 67

EP - 80

JO - Neural networks

JF - Neural networks

SN - 0893-6080

IS - January

ER -