And here's another recent article from Democritus University of Thrace examining using spiking neural networks and STDP in an unconventional approach. It uses less neurons and synapses than other approaches and yet when applied to a MNIST dataset still achieves reasonably high accuracy. This is another approach where neuromorphic hardware like Akida would benefit.
FYI it's quite a heavy read.
https://arxiv.org/pdf/2008.03330.pdf
2021 BRN Discussion, page-5834
-
- There are more pages in this discussion • 27,747 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Add BRN (ASX) to my watchlist
 (20min delay)
|
|||||
|
Last
15.5¢ |
Change
-0.005(3.13%) |
Mkt cap ! $304.1M | |||
| Open | High | Low | Value | Volume |
| 16.5¢ | 16.5¢ | 15.5¢ | $537.1K | 3.385M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 55 | 1835555 | 15.5¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 16.0¢ | 230384 | 13 |
View Market Depth
| Last trade - 16.10pm 10/09/2024 (20 minute delay) ? |
| BRN (ASX) Chart |