Intrinsic Robotic Introspection: Learning Internal States From Neuron Activations

Pitsillos, N., Pore, A., Jensen, B. S. and Aragon Camarasa, G. (2021) Intrinsic Robotic Introspection: Learning Internal States From Neuron Activations. In: IEEE International Conference on Development and Learning (ICDL 2021), Beijing, China, 23-26 Aug 2021, ISBN 9781728162423 (doi:10.1109/ICDL49984.2021.9515672)

[img] Text
245283.pdf - Accepted Version

923kB

Abstract

We present an introspective framework inspired by the process of how humans perform introspection. Our working assumption is that neural network activations encode information, and building internal states from these activations can improve the performance of an actor-critic model. We perform experiments where we first train a Variational Autoencoder model to reconstruct the activations of a feature extraction network and use the latent space to improve the performance of an actor-critic when deciding which low-level robotic behaviour to execute. We show that internal states reduce the number of episodes needed by about 1300 episodes while training an actor-critic, denoting faster convergence to get a high success value while completing a robotic task.

Item Type:Conference Proceedings
Additional Information:This research has been supported by EPSRC DTA No. 2279292 and NVIDIA Corporation for the donation of the Titan Xp GPU.
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Aragon Camarasa, Dr Gerardo and Pitsillos, Nikos and Jensen, Dr Bjorn
Authors: Pitsillos, N., Pore, A., Jensen, B. S., and Aragon Camarasa, G.
College/School:College of Science and Engineering
College of Science and Engineering > School of Computing Science
ISBN:9781728162423
Published Online:20 August 2021
Copyright Holders:Copyright Copyright © 2021, IEEE
Publisher Policy:Reproduced in accordance with the publisher copyright policy
Related URLs:

University Staff: Request a correction | Enlighten Editors: Update this record