I prefer not to indulge in dialogue with wishful thinkers, it is generally an exercise in futility. I did a MSc in Ai but i don't pretend to be au fait with the lastest nuances in SNNs though i can tell you that was passes for research on this thread is generally risible, with hype being routinely mistaken for fact. SNNs in their current form are not well regarded by a lot of people in the industry. Yann Le Cun, Facebook's head of AI, and a formidable opinion leader, was expresssing his contempt for them just the other day, and even a paper last year. cited on here, by some of the inventors of JAST admitted that SNNs still cannot match CNNs for either efficiency OR performance - a bit of a double whammy, though it spoke coyly of 'hopeful signs' which an optimist might hope referred to their own efforts.
Here is a link about the SNN CNN debate that is written by someone who actually does know what they are talking about, you can see - if your blinkers permit it that it explains and effectively predicts (it is from 2017) the difficulties BRN have encountered far better than anyone here has ever managed. Strangely enough no-one seems to have turned it up as part of their 'research', except me and I was reluctant to post it in the face of all the unmitigated optimism.
Deep Learning or Obscurity?
Th fact that some form of neurmorphic computing more closely modelled on the brain is the medium term future does not mean current efforts like BRN's are going to suceed, the field still needs further fundamental breakthroughs. The fact that current models are so focussed on mimicking deep learning is telling you by itself that they have significant shortcomings.
The most alarming thing about BRN remiais the lack of quantifiable user interest, other AI co's like rival start Chip Maker Graphcore are innundated with interest and have Microsoft BMW etc queueing up to take a stake, likewise Blue Prism another AI holding of mine has hundreds of projects on the go and peaked at 30times its listing price.
That said I expect a bounce here soon-ish, it is an AI co and everyone wants to be in AI, if they have to do a retail raise at 5-7 cents or below it will be a grim and depressing indictment of management, money (it goes with out saying at a higher level) from some kind of partnership or specialist fund would be much preferable, it's not as if the precedents aren't there with their peers.
- Forums
- ASX - By Stock
- BRN
- Brainchip's Akida - what does it even do?
Brainchip's Akida - what does it even do?, page-26
-
- There are more pages in this discussion • 25 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Featured News
Add BRN (ASX) to my watchlist
|
|||||
Last
16.0¢ |
Change
0.005(3.23%) |
Mkt cap ! $304.1M |
Open | High | Low | Value | Volume |
15.5¢ | 16.0¢ | 15.5¢ | $298.6K | 1.920M |
Buyers (Bids)
No. | Vol. | Price($) |
---|---|---|
53 | 1362115 | 15.5¢ |
Sellers (Offers)
Price($) | Vol. | No. |
---|---|---|
16.0¢ | 1156375 | 41 |
View Market Depth
No. | Vol. | Price($) |
---|---|---|
53 | 1362115 | 0.155 |
138 | 4582061 | 0.150 |
42 | 981106 | 0.145 |
44 | 1054329 | 0.140 |
8 | 269794 | 0.135 |
Price($) | Vol. | No. |
---|---|---|
0.160 | 1156375 | 41 |
0.165 | 912911 | 17 |
0.170 | 928486 | 17 |
0.175 | 887112 | 13 |
0.180 | 1345875 | 18 |
Last trade - 15.45pm 06/09/2024 (20 minute delay) ? |
Featured News
BRN (ASX) Chart |