BRN Discussion Ongoing


Bit more info on the experiment


And if you scroll down to the bottom, I recon someone could come up with a good use case and reach out to them


If you have ideas for other use-cases or product designs using the Brainchip Akida neuromorphic processor, be sure to reach out to us!
 
  • Like
  • Love
  • Fire
Reactions: 20 users
Not new but now showing Akida 1000 in the photo

See that Numem also there. Link to CEO Numem LinkedIn page. Notice the companies he mentions.

https://www.linkedin.com/posts/acti...3q?utm_source=share&utm_medium=member_desktop

SC
 
  • Like
  • Fire
Reactions: 12 users

AARONASX

Holding onto what I've got
  • Like
  • Fire
Reactions: 3 users

Boab

I wish I could paint like Vincent
  • Like
  • Love
Reactions: 27 users

hotty4040

Regular
Extinction list -

Which of the following will not partner with BrainChip:

ARM,
Intel,
Nvidia,
Qualcomm,
Amazon,
Google,
IBM,
AMD,
TI,
... ?

Good question Dio
 
  • Like
Reactions: 1 users

Beebo

Regular
Extinction list -

Which of the following will not partner with BrainChip:

ARM,
Intel,
Nvidia,
Qualcomm,
Amazon,
Google,
IBM,
AMD,
TI,
... ?
Curious as to your best guess!!!
 
  • Like
Reactions: 1 users

stuart888

Regular

Interesting that you/me/anyone can use AI to create a video like this. Ask your favorite ChatBot to give a 70 second summary of Brainchip, with a dynamic image creator, and AI voice output.

Boom, easy video content, to make money 💰 on Youtube!
And you promote Brainchip and help your Stock Portfolio at the same time!!!
 
  • Like
  • Fire
Reactions: 12 users

stuart888

Regular
ViT is a new feature of Akida 2, but, according to Prof. Wiki, the idea of transformers has been around for several years, although the concept was applied to vision a couple of years ago.
Hey Google: "vit vision transformer simplified"? Patches?

1711666559035.png

Here's how ViTs work:
  1. ViTs break down an image into patches
  2. Each patch is serialized into a vector
  3. The vector is mapped to a smaller dimension with a single matrix multiplication
  4. The output from each patch's self-attention mechanism is passed through a feedforward neural network
  5. The feedforward network introduces non-linearity and allows the model to learn complex relationships between patches
Clear as mud!
 
  • Like
  • Haha
Reactions: 12 users

stuart888

Regular
As a footnote to my early Monday morning post regards the Accenture Global Solutions Patent Application I suggest following Tim Shea whom I have been observing on Linkedin for at least 6 or more years, he is a key player in Neuromorphic Computing, having been with Accenture and now tight with Mike Davies and Intel Labs...very cleaver man.
Just my view nothing more at this point.
Cheers Tech. ♦️
Intel is all in for Neuromorphic for sure. They spent some money on this video!
 
  • Like
  • Fire
  • Love
Reactions: 20 users

IloveLamp

Top 20
Each sensor, measuring 300 by 300 micrometers, acts as a wireless node in a large array, analogous to neurons in the brain. When a node senses an event, such as a change in temperature or neural activity, the device sends the data as a “spike” signal, consisting of a series of short radiofrequency pulses, to a central receiver. That receiver then decodes the information.

Corradi believes this work is part of a larger trend toward neuromorphic systems, a “new wave of brain-machine interfaces that I hope we will see in the coming future.”

 
  • Like
  • Love
  • Wow
Reactions: 15 users

stuart888

Regular
Each sensor, measuring 300 by 300 micrometers, acts as a wireless node in a large array, analogous to neurons in the brain. When a node senses an event, such as a change in temperature or neural activity, the device sends the data as a “spike” signal, consisting of a series of short radiofrequency pulses, to a central receiver. That receiver then decodes the information.

Corradi believes this work is part of a larger trend toward neuromorphic systems, a “new wave of brain-machine interfaces that I hope we will see in the coming future.”

While the technology could be applied to any part of the body where biomedical researchers aim to monitor physiological activity, the primary goal is use in a brain-machine interface that can probe a large region of the brain, says Nurmikko. The sensors could also be modified for use in wearable technology or environmental sensors.

There are key advantages of the system for biomedical uses, such as the small, unobtrusive design. But these applications also impose a key limitation: The sensors are externally powered by a wireless beam to avoid the need for batteries, and the body can only safely absorb so much radiofrequency energy. In other words, the system is not limited by bandwidth, but instead by power delivery. “From a practical point of view, it always comes back to the question of, where do you get your energy?” says Nurmikko.

One important element of the Brown researcher’s design is its simplicity, says Corradi. The sensor’s architecture does not include a battery or clock embedded within the chips, making it ideal for scalable, low-power systems. “It opens a lot of possibilities.”
 
  • Like
Reactions: 9 users
Intel is all in for Neuromorphic for sure. They spent some money on this video!

He forgot to mention they have partnered with the best outside help Available brainchip to get them across the line at the edge . I mean to say Mike, if someone’s giving you a piggyback and your winning the race wouldn’t you think of given the bloke carrying you a shoutout for respect sake at least 😆
Go Brainchip.
 
Last edited:
  • Fire
  • Haha
  • Like
Reactions: 12 users
Good read here. Not sure if Brainchip is involved in all of this, but great to see where things are heading:


"Microsoft’s Copilot will be able to run locally on your AI PC in the future, it has been confirmed, as have the hardware requirements in terms of how powerful an NPU (Neural Processing Unit) will be required in these AI-focused devices."

"As mentioned, it was further confirmed that in the future Microsoft’s Copilot AI will be able to run locally on the AI PC – which this additional NPU processing power will facilitate, as opposed to needing to be online and tap the power of the cloud for the AI assistant’s responses. Or at least some, or perhaps a great deal, of Copilot functionality will be handled locally."

"Lewellen clarified: “And as we go to that next gen [AI PC], it’s just going to enable us to run more things locally, just like they will run Copilot with more elements of Copilot running locally on the client. That may not mean that everything in Copilot is running local, but you’ll get a lot of key capabilities that will show up running on the NPU.”
 
  • Like
  • Love
  • Fire
Reactions: 11 users

7für7

Regular
Thanks FF there is nothing more to add on this! Agree 100%! Have a nice eastern time everyone

“I have been invested now for over seven years. It was a highly speculative investment when I first entered and one which was wholly valued by reference to AKIDA technology potential. In other words it was a research company with huge ambition.

I was invested when the now infamous words by an investment advisor were published "What's a neuromorphic chip anyway. The way I see it if it was any good it would not be on the ASX."

The mentally challenged however still cling to this now defeated argument that the Brainchip AKIDA technology is not as described by the founders of Brainchip and is not as described by Edge Impulse as being "Science Fiction" made real nor is it as described in a Forbes article as mind boggling technology.

The independent and prestigious endorsements of AKIDA technology are many and numerous and coming thick and fast yet this mentally challenged group cannot adjust their arguments to something which is even remotely intelligent despite being confronted with one of the most respected automotive manufacturers in the world stating publicly that AKIDA was five to ten times more efficient than competitors.

All this before they are suffocated by Tata Consulting Services multiple research papers proving out the superiority of AKIDA technology to the best Nvidia has to offer.

Only today I was reading a research paper which was comparing Loihi from Intel to a range of other accelerators and in the conclusion it states:

CONCLUSION

In this paper, we focused on the development and optimization of ML models designed for performing facial expression recognition at the edge. In the first phase of the research, we leveraged hardware-aware neural architecture search to optimize CNN models for deployment on four distinct edge AI accelerators. Among these edge platforms, the coral Dev Board emerged as the best option offering the lowest latency, power dissipation, and energy consumption while maintaining a high level of accuracy. In the second phase, we delved into the conversion of the CNN models into SNNs for deployment on the Loihi neuromorphic processor. To enhance the models’ performance within the neuromorphic system, we explored the impact of increasing input image sparsity through an additional pre-processing step involving edge detection. Our findings revealed that edge detection not only reduced power dissipation and energy consumption by decreasing the neural activities within SNN layers but also led to improved accuracy which can be attributed to less noisy inputs. Ultimately, we compared Loihi against edge AI accelerators in terms of accuracy, latency, power, and energy consumption. The comparison results underscored Loihi’s remarkable advantages, with approximately two orders of magnitude reduction in power dissipation and one order of magnitude in energy savings compared to edge AI accelerators while maintaining a comparable accuracy. In terms of latency, Loihi lagged behind the edge accelerators. Nonetheless, it managed to fulfill the real-time requirements with a processing speed of more than 28 frames per second.


https://arxiv.org/pdf/2403.08792

I do not intend to extract the document as it is on the Brainchip website showing that Nviso using AKD1000 for analysing facial expression was able to take AKIDA to over 1,000 frames per second at less than $25.00 a pop.

Is it any wonder that Intel has partnered with Brainchip and promotes it centre stage during Pat Gelsinger's keynote and why Mike Davies has moved Loihi research towards data centre applications away from the Edge and suggests that a commercial Loihi is up to five years away.

Then to nail the coffin even more tightly closed on this lost and futile argument Accenture came out in favour of AKIDA technology brilliance in the last week.

There is no intellectual credibility to be had or earned by making derogatory statements about the value of AKIDA technology.

That train has long since left the station.

The clue for all those who lack the intellect to come up with other ideas with which to attack Brainchip is to confine attacks to management and lack of commercial success. If they do so, while they will not necessarily be successful, they will not immediately be exposed as coming to the gun fight with an empty holster.

A further tip they might like to take on board is that they can probably still run the argument that Brainchip does not even have a product in the market. They will need to be quick however as we will shortly tick over into April and VVDN AKIDA Edge Box and Unigen AKIDA Edge Server are ever so close to being shipped.

Also they should be careful when attacking the commercial relationship Brainchip has with MicroChip because at CES 2024 in the Brainchip podcast the MicroChip representative stated they were now ready to take their joint offering with Brainchip to their customers.

The true investor shareholders here know there are many more partnerships and engagements publicly revealed running to at least sixty to consider and involving companies like ARM, Prophesee, Tata Elxsi, SiFive, Quantum Ventura, Renesas, MegaChips, Moschip, NASA, EDGX, Valeo. Infineon, OnSemiconductor and the like. I even published my list recently which an intelligent manipulator could dig into and try and find material to attack Brainchip over delays.

I suspect that this approach is beyond most of their capabilities and the remainder know that sitting out there like the elephant in the room is the fact that Prophesee and Sony first announced they were partnering in 2020. It was clear from their announcement that they had been engaged for some time before the announcement and it was only recently that Prophesee announced when they expected their joint product offering to become available. Prophesee like Brainchip is bringing revolutionary technology to market and stands as a comparable as to how long such an endeavour takes.

My recommendation to every true investor shareholder is to simply place on ignore those who now make claims derogatory of the AKIDA technology as their actual intent is immediately clear and any new investor will with even the most superficial of research realise quickly that such statements have no value and should be ignored.

My opinion only DYOR

Fact Finder”
 
Last edited:
  • Like
  • Fire
  • Love
Reactions: 28 users
  • Haha
  • Like
Reactions: 4 users

Rskiff

Regular
Thanks FF there is nothing more to add on this! Agree 100%! Have a nice eastern time everyone

what does the link say as I refuse to click on that site and help up the post count for them.
 
  • Like
  • Fire
  • Haha
Reactions: 8 users

Esq.111

Fascinatingly Intuitive.
Chippers ,

Revised share price.

Regards,
Esq.
 

Attachments

  • 20240329_161754.jpg
    20240329_161754.jpg
    1.9 MB · Views: 209
  • Like
  • Love
  • Fire
Reactions: 37 users

KKFoo

Regular
Good afternoon Chippers, I have a question, in the 2023 yearly report it was reported that the revenue was around 230k, does it means that this coming 4 c we won't expect the cash receipt to be anything more than the 230k (assuming that those 230k has not been paid yet)?
 
Top Bottom