Cargando…
Neuromorphic Engineering Needs Closed-Loop Benchmarks
Neuromorphic engineering aims to build (autonomous) systems by mimicking biological systems. It is motivated by the observation that biological organisms—from algae to primates—excel in sensing their environment, reacting promptly to their perils and opportunities. Furthermore, they do so more resil...
Autores principales: | , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8884247/ https://www.ncbi.nlm.nih.gov/pubmed/35237122 http://dx.doi.org/10.3389/fnins.2022.813555 |
_version_ | 1784660109595508736 |
---|---|
author | Milde, Moritz B. Afshar, Saeed Xu, Ying Marcireau, Alexandre Joubert, Damien Ramesh, Bharath Bethi, Yeshwanth Ralph, Nicholas O. El Arja, Sami Dennler, Nik van Schaik, André Cohen, Gregory |
author_facet | Milde, Moritz B. Afshar, Saeed Xu, Ying Marcireau, Alexandre Joubert, Damien Ramesh, Bharath Bethi, Yeshwanth Ralph, Nicholas O. El Arja, Sami Dennler, Nik van Schaik, André Cohen, Gregory |
author_sort | Milde, Moritz B. |
collection | PubMed |
description | Neuromorphic engineering aims to build (autonomous) systems by mimicking biological systems. It is motivated by the observation that biological organisms—from algae to primates—excel in sensing their environment, reacting promptly to their perils and opportunities. Furthermore, they do so more resiliently than our most advanced machines, at a fraction of the power consumption. It follows that the performance of neuromorphic systems should be evaluated in terms of real-time operation, power consumption, and resiliency to real-world perturbations and noise using task-relevant evaluation metrics. Yet, following in the footsteps of conventional machine learning, most neuromorphic benchmarks rely on recorded datasets that foster sensing accuracy as the primary measure for performance. Sensing accuracy is but an arbitrary proxy for the actual system's goal—taking a good decision in a timely manner. Moreover, static datasets hinder our ability to study and compare closed-loop sensing and control strategies that are central to survival for biological organisms. This article makes the case for a renewed focus on closed-loop benchmarks involving real-world tasks. Such benchmarks will be crucial in developing and progressing neuromorphic Intelligence. The shift towards dynamic real-world benchmarking tasks should usher in richer, more resilient, and robust artificially intelligent systems in the future. |
format | Online Article Text |
id | pubmed-8884247 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2022 |
publisher | Frontiers Media S.A. |
record_format | MEDLINE/PubMed |
spelling | pubmed-88842472022-03-01 Neuromorphic Engineering Needs Closed-Loop Benchmarks Milde, Moritz B. Afshar, Saeed Xu, Ying Marcireau, Alexandre Joubert, Damien Ramesh, Bharath Bethi, Yeshwanth Ralph, Nicholas O. El Arja, Sami Dennler, Nik van Schaik, André Cohen, Gregory Front Neurosci Neuroscience Neuromorphic engineering aims to build (autonomous) systems by mimicking biological systems. It is motivated by the observation that biological organisms—from algae to primates—excel in sensing their environment, reacting promptly to their perils and opportunities. Furthermore, they do so more resiliently than our most advanced machines, at a fraction of the power consumption. It follows that the performance of neuromorphic systems should be evaluated in terms of real-time operation, power consumption, and resiliency to real-world perturbations and noise using task-relevant evaluation metrics. Yet, following in the footsteps of conventional machine learning, most neuromorphic benchmarks rely on recorded datasets that foster sensing accuracy as the primary measure for performance. Sensing accuracy is but an arbitrary proxy for the actual system's goal—taking a good decision in a timely manner. Moreover, static datasets hinder our ability to study and compare closed-loop sensing and control strategies that are central to survival for biological organisms. This article makes the case for a renewed focus on closed-loop benchmarks involving real-world tasks. Such benchmarks will be crucial in developing and progressing neuromorphic Intelligence. The shift towards dynamic real-world benchmarking tasks should usher in richer, more resilient, and robust artificially intelligent systems in the future. Frontiers Media S.A. 2022-02-14 /pmc/articles/PMC8884247/ /pubmed/35237122 http://dx.doi.org/10.3389/fnins.2022.813555 Text en Copyright © 2022 Milde, Afshar, Xu, Marcireau, Joubert, Ramesh, Bethi, Ralph, El Arja, Dennler, van Schaik and Cohen. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. |
spellingShingle | Neuroscience Milde, Moritz B. Afshar, Saeed Xu, Ying Marcireau, Alexandre Joubert, Damien Ramesh, Bharath Bethi, Yeshwanth Ralph, Nicholas O. El Arja, Sami Dennler, Nik van Schaik, André Cohen, Gregory Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title | Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title_full | Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title_fullStr | Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title_full_unstemmed | Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title_short | Neuromorphic Engineering Needs Closed-Loop Benchmarks |
title_sort | neuromorphic engineering needs closed-loop benchmarks |
topic | Neuroscience |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8884247/ https://www.ncbi.nlm.nih.gov/pubmed/35237122 http://dx.doi.org/10.3389/fnins.2022.813555 |
work_keys_str_mv | AT mildemoritzb neuromorphicengineeringneedsclosedloopbenchmarks AT afsharsaeed neuromorphicengineeringneedsclosedloopbenchmarks AT xuying neuromorphicengineeringneedsclosedloopbenchmarks AT marcireaualexandre neuromorphicengineeringneedsclosedloopbenchmarks AT joubertdamien neuromorphicengineeringneedsclosedloopbenchmarks AT rameshbharath neuromorphicengineeringneedsclosedloopbenchmarks AT bethiyeshwanth neuromorphicengineeringneedsclosedloopbenchmarks AT ralphnicholaso neuromorphicengineeringneedsclosedloopbenchmarks AT elarjasami neuromorphicengineeringneedsclosedloopbenchmarks AT dennlernik neuromorphicengineeringneedsclosedloopbenchmarks AT vanschaikandre neuromorphicengineeringneedsclosedloopbenchmarks AT cohengregory neuromorphicengineeringneedsclosedloopbenchmarks |