| Téléchargement | - Voir la version finale : Comparing Loihi with a SpiNNaker 2 prototype on low-latency keyword spotting and adaptive robotic control (PDF, 3.2 Mio)
|
|---|
| DOI | Trouver le DOI : https://doi.org/10.1088/2634-4386/abf150 |
|---|
| Auteur | Rechercher : Yan, YexinIdentifiant ORCID : https://orcid.org/0000-0002-4267-7015; Rechercher : Stewart, Terrence C.1; Rechercher : Choo, Xuan; Rechercher : Vogginger, Bernhard; Rechercher : Partzsch, Johannes; Rechercher : Höppner, Sebastian; Rechercher : Kelber, Florian; Rechercher : Eliasmith, Chris; Rechercher : Furber, Steve; Rechercher : Mayr, Christian |
|---|
| Affiliation | - Conseil national de recherches du Canada. Technologies numériques
|
|---|
| Bailleur de fonds | Rechercher : National Research Council of Canada; Rechercher : Horizon 2020 Framework Programme; Rechercher : Intel Corporation; Rechercher : Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Rechercher : Seventh Framework Programme |
|---|
| Format | Texte, Article |
|---|
| Sujet | SpiNNaker; MAC array; Loihi; neuromorphic computing; adaptive robotic control; keyword spotting |
|---|
| Résumé | Abstract We implemented two neural network based benchmark tasks on a prototype chip of the second-generation SpiNNaker (SpiNNaker 2) neuromorphic system: keyword spotting and adaptive robotic control. Keyword spotting is commonly used in smart speakers to listen for wake words, and adaptive control is used in robotic applications to adapt to unknown dynamics in an online fashion. We highlight the benefit of a multiply-accumulate (MAC) array in the SpiNNaker 2 prototype which is ordinarily used in rate-based machine learning networks when employed in a neuromorphic, spiking context. In addition, the same benchmark tasks have been implemented on the Loihi neuromorphic chip, giving a side-by-side comparison regarding power consumption and computation time. While Loihi shows better efficiency when less complicated vector-matrix multiplication is involved, with the MAC array, the SpiNNaker 2 prototype shows better efficiency when high dimensional vector-matrix multiplication is involved. |
|---|
| Date de publication | 2021-07-15 |
|---|
| Maison d’édition | IOP Publishing |
|---|
| Licence | |
|---|
| Dans | |
|---|
| Langue | anglais |
|---|
| Publications évaluées par des pairs | Oui |
|---|
| Exporter la notice | Exporter en format RIS |
|---|
| Signaler une correction | Signaler une correction (s'ouvre dans un nouvel onglet) |
|---|
| Identificateur de l’enregistrement | 067f85d0-4a8f-4c0f-894b-1997b2bcb569 |
|---|
| Enregistrement créé | 2023-09-18 |
|---|
| Enregistrement modifié | 2023-09-18 |
|---|
