Benedict Jones benedict.jones@durham.ac.uk
PGR Student Doctor of Philosophy
In-Materio Extreme Learning Machines
Jones, Benedict A.H.; Al Moubayed, Noura; Zeze, Dagou A.; Groves, Chris
Authors
Dr Noura Al Moubayed noura.al-moubayed@durham.ac.uk
Associate Professor
Professor Dagou Zeze d.a.zeze@durham.ac.uk
Professor
Professor Chris Groves chris.groves@durham.ac.uk
Professor
Contributors
Günter Rudolph
Editor
Anna V. Kononova
Editor
Hernán Aguirre
Editor
Pascal Kerschke
Editor
Gabriela Ochoa
Editor
Tea Tušar
Editor
Abstract
Nanomaterial networks have been presented as a building block for unconventional in-Materio processors. Evolution in-Materio (EiM) has previously presented a way to congure and exploit physical materials for computation, but their ability to scale as datasets get larger and more complex remains unclear. Extreme Learning Machines (ELMs) seek to exploit a randomly initialised single layer feed forward neural network by training the output layer only. An analogy for a physical ELM is produced by exploiting nanomaterial networks as material neurons within the hidden layer. Circuit simulations are used to eciently investigate diode-resistor networks which act as our material neurons. These in-Materio ELMs (iM-ELMs) outperform common classication methods and traditional articial ELMs of a similar hidden layer size. For iM-ELMs using the same number of hidden layer neurons, leveraging larger more complex material neuron topologies (with more nodes/electrodes) leads to better performance, showing that these larger materials have a better capability to process data. Finally, iM-ELMs using virtual material neurons, where a single material is re-used as several virtual neurons, were found to achieve comparable results to iM-ELMs which exploited several dierent materials. However, while these Virtual iM-ELMs provide signicant exibility, they sacrice the highly parallelised nature of physically implemented iM-ELMs.
Citation
Jones, B. A., Al Moubayed, N., Zeze, D. A., & Groves, C. (2022). In-Materio Extreme Learning Machines. In G. Rudolph, A. V. Kononova, H. Aguirre, P. Kerschke, G. Ochoa, & T. Tušar (Eds.), Parallel Problem Solving from Nature – PPSN XVII (505-519). Springer Verlag. https://doi.org/10.1007/978-3-031-14714-2_35
Acceptance Date | Jun 6, 2022 |
---|---|
Online Publication Date | Aug 14, 2022 |
Publication Date | 2022 |
Deposit Date | Aug 3, 2022 |
Publicly Available Date | Aug 15, 2023 |
Publisher | Springer Verlag |
Pages | 505-519 |
Series Title | Lecture Notes in Computer Science |
Series Number | 13398 |
Book Title | Parallel Problem Solving from Nature – PPSN XVII |
ISBN | 978-3-031-14713-5 |
DOI | https://doi.org/10.1007/978-3-031-14714-2_35 |
Public URL | https://durham-repository.worktribe.com/output/1649976 |
Files
Accepted Book Chapter
(3.6 Mb)
PDF
Copyright Statement
The final authenticated version is available online at https://doi.org/10.1007/978-3-031-14714-2_35
You might also like
Is Unimodal Bias Always Bad for Visual Question Answering? A Medical Domain Study with Dynamic Attention
(2022)
Conference Proceeding
Towards Graph Representation Learning Based Surgical Workflow Anticipation
(2022)
Conference Proceeding
Does lossy image compression affect racial bias within face recognition?
(2022)
Conference Proceeding
Downloadable Citations
About Durham Research Online (DRO)
Administrator e-mail: dro.admin@durham.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search