| Download | - View final version: Comparing Loihi with a SpiNNaker 2 prototype on low-latency keyword spotting and adaptive robotic control (PDF, 3.2 MiB)
|
|---|
| DOI | Resolve DOI: https://doi.org/10.1088/2634-4386/abf150 |
|---|
| Author | Search for: Yan, YexinORCID identifier: https://orcid.org/0000-0002-4267-7015; Search for: Stewart, Terrence C.1; Search for: Choo, Xuan; Search for: Vogginger, Bernhard; Search for: Partzsch, Johannes; Search for: Höppner, Sebastian; Search for: Kelber, Florian; Search for: Eliasmith, Chris; Search for: Furber, Steve; Search for: Mayr, Christian |
|---|
| Name affiliation | - National Research Council of Canada. Digital Technologies
|
|---|
| Funder | Search for: National Research Council of Canada; Search for: Horizon 2020 Framework Programme; Search for: Intel Corporation; Search for: Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada; Search for: Seventh Framework Programme |
|---|
| Format | Text, Article |
|---|
| Subject | SpiNNaker; MAC array; Loihi; neuromorphic computing; adaptive robotic control; keyword spotting |
|---|
| Abstract | 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. |
|---|
| Publication date | 2021-07-15 |
|---|
| Publisher | IOP Publishing |
|---|
| Licence | |
|---|
| In | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | 067f85d0-4a8f-4c0f-894b-1997b2bcb569 |
|---|
| Record created | 2023-09-18 |
|---|
| Record modified | 2023-09-18 |
|---|
