BRN Discussion Ongoing

[Diogenese]

Hi SFB,

Great find:

Top right "NNA" = neural network accelerator - Akida baked into their processor!

... and immediately under NNA, "User defined Logic" allows customers to customize the processor.

View attachment 26956

Advanced AI Solutions for Automotive​

Socionext has partnered with artificial intelligence provider BrainChip to develop optimized, intelligent sensor data solutions based on Brainchip’s Akida® processor IP.

BrainChip’s flexible AI processing fabric IP delivers neuromorphic, event-based computation, enabling ultimate performance while minimizing silicon footprint and power consumption. Sensor data can be analyzed in real-time with distributed, high-performance and low-power edge inferencing, resulting in improved response time and reduced energy consumption.

How good is that?!

Had a look at Socionext artificial intelligence/neural network patents.

Seems they had not progressed much beyond von Neumann CNNs (software).

...

and then, along came Akida ...


When Anil approached them to make the first Akida chips, now that they have seen the finished silicon product, they must have gone from incredulity to astonishment.

So now they have a complete CPU/GPU/NN customizable design ready for customers to put their finishing touches on, nodoubt along with a couple they'd baked earlier.

The customizable part could possibly be in the form of a FPGA, so if you can wait 5 minutes, we'll just pop it in the oven to warm it up.

 

[buena suerte :-)]

Study Finds 100% Of Men Would Eat Any Fruit Given To Them By A Naked Woman​

WEST OF EDEN - A double-blind study conducted by angel scientists has confirmed that 100% of men would eat any fruit given to them by a naked woman. The study seems to suggest that the fall of Adam and Eve was unavoidable.

"Our results were conclusive," said the angel Gabriel, who led the study. "Of the 30,000 male test subjects we studied, we found literally 100% of them would happily accept and eat any piece of fruit handed to them by a woman who was naked. Even in cases where it wasn't fruit - or even edible - all the males gratefully accepted every single object offered and ate it with a dumb look on their face."

Gabriel then motioned to a male test subject on the other side of the glass as he happily chewed on a rubber ball that had just been handed to him by a gorgeous unclothed woman.

"See what I mean?"

Researchers say the results should serve to humble any man who thinks he would have made a different decision than Adam. "According to the science, you would have likely done the exact same thing Adam did," said Gabriel.

Women around the world started to gloat upon hearing the news until Gabriel revealed they also conducted a study showing that 100% of women can be tricked by a talking snake.

I think that should be set up as a new thread called 'The Friday Funny' just to start the weekend off

 

[David holland]

S&P 500 is actually a great way of following markets

And I agree ignore is a great function to use.

Not sure why FF was feeling the angst, we are all brainchip family here, if he is taking a break I completely understand.
It's hard to keep supporting some people who are always anxious when share price goes down and throws a tantrum for no reason. I do empathise with them though, as everyone deals with it differently.

Come on they talk the CES in Vegas, Everybody is going to be talking Brainchip Shareprice goes up then crashes, then a podcast comes out and everybody is excited again, Partnerships are formed and Everybody talks a massive game but Shareprice is in the 60 's , yeah I can understand the frustration

 

[Rise from the ashes]

What's your thoughts as a longtime shareholder has your confidence in BRN grown over the last couple of years?

 

[thelittleshort]

 

[IloveLamp]

No issues promoting Brainchip in a respectful way. There are a few posters who do it brilliantly too.

Agree with not being arrogant to other company personnel, especially when Intel is our partner. It's a great opening for Brainchip to go next level.
US government is supporting Intel heavily for local businesses initiative, let's not kill the golden goose.

[Video] Eric McLaughlin on LinkedIn: #virtualreality #oculusquest #13thgen #vr #accesspoint #iamintel #wifi…

#virtualreality experiences are now faster and superior through the collaboration between Meta #oculusquest devices and Intel Corporation Double Connect…

www.linkedin.com

Brad Aimone on LinkedIn: A computing hardware approach aspires to emulate the brain | 10 comments

This is an exciting article in Physics Today talking about, among other things, the prospects of large-scale neuromorphic platforms like SpiNNaker and Loihi to… | 10 comments on LinkedIn

www.linkedin.com

 

[JB49]

Interesting lecture here presented by Dr Heinrihc Gotzig explaining how the Valeo radars work. He appears to be one of the experts at Valeo.

Even though the brainchip website lists that we are "trusted by" valeo, I'm always very pessimistic until I see it written in concrete! For all I know, they could be trusting us in a research capacity rather than using brainchip in their products. Nonetheless, I'm very hopeful!!

Throughout the lecture he mainly refers to CNN's. But at around the 47.15 minute mark, he does refer to Spiking Neural Networks, but only that he is currently running a separate PHD on it. Interesting comment, but couldnt find anything in the video giving away that they are using akida.

 

[JB49]

He's also written the following research paper on spiking neural networks.

https://arxiv.org/pdf/1903.02080v1.pdf

 

[Boab]

Perhaps these huge screens/dashboards will disappear due to safety concerns??

The BMW boss made the comments as the German car-maker unveiled a potential solution to the problem – a full-width head-up display system which extends across the the windscreen.

Big infotainment screens in cars won't last – BMW boss predicts safety backlash

The global boss of BMW believes giant in-car infotainment screens could disappear within a decade due to their potential to distract drivers.

www.drive.com.au

 

[TheDrooben]

Cyber-Neuro RT: Real-time Neuromorphic Cybersecurity​

Cyber-Neuro RT: Real-time Neuromorphic Cybersecurity

High throughput environments, such as those found in high performance computing (HPC) clusters, run at substantially greater scales than standard busi…

reader.elsevier.com

Comparison study between Loihi and Akida..............no wonder Intel moved on from Loihi and agreed to bring Akida into their IFS

Larry

 

[Boab]

Perhaps these huge screens/dashboards will disappear due to safety concerns??

The BMW boss made the comments as the German car-maker unveiled a potential solution to the problem – a full-width head-up display system which extends across the the windscreen.

Big infotainment screens in cars won't last – BMW boss predicts safety backlash

The global boss of BMW believes giant in-car infotainment screens could disappear within a decade due to their potential to distract drivers.

www.drive.com.au

After further thought......if we can move to full autonomy we can keep the big screens as we can look away.

 

[JDelekto]

No issues promoting Brainchip in a respectful way. There are a few posters who do it brilliantly too.

Agree with not being arrogant to other company personnel, especially when Intel is our partner. It's a great opening for Brainchip to go next level.
US government is supporting Intel heavily for local businesses initiative, let's not kill the golden goose.

Like others here, I cringe when people use a certain tact of advertising Akida either on social media or other forums, especially when the attempt is under the guise of a naive question.

What I do recommend, for those who frequent sites like LinkedIn, is to add emotes/likes to BrainChip announcements.

Likewise, if one sees an employee, partner, or even competitor from another company providing positive words or praise for BrainChip's technology, I would recommend giving that positive feedback as well.

 

[Diogenese]

Cyber-Neuro RT: Real-time Neuromorphic Cybersecurity​

Cyber-Neuro RT: Real-time Neuromorphic Cybersecurity

High throughput environments, such as those found in high performance computing (HPC) clusters, run at substantially greater scales than standard busi…

reader.elsevier.com

Comparison study between Loihi and Akida..............no wonder Intel moved on from Loihi and agreed to bring Akida into their IFS

View attachment 26988




Larry

Hi Larry,

I don't think the times really tell us much.

The test was done using the Akida ADE simulation software:

"3.3. Intel Loihi and BrainChip Akida
The 10-layer deep learning network was converted into formats compatible with Intel and BrainChip. In both cases, precision of the weights was varied to see how it affected model size and accuracy. Timing results were also produced.
For Intel, we accessed Loihi (Nahuku32) chips via the cloud and NxSDK 1.0;
for BrainChip, we simulated the chips on CPU via their hardware abstraction layer. Timing results are not comparable because CPU or GPU simulations are much faster than running on neuromorphic hardware."

5.
"Due to this constraint, our experiments were conducted using Intel’s cloud environment and the stand-alone simulation environment for Akida. There are drastic differences between software simulation and hardware and we intend to study those further."

However, this does explain something else. I saw a quote a couple of weeks ago which referred to 8-bit akida, and, as we all know, Akida only has 4 bits. However, apparently when using the Akida simulator, you can choose 8 bits.

"For BrainChip, accuracy also increased with precision; maximum accuracy was 82.5% at 8 bits and 66.8% at 4 bits. The 8-bit model produced results in 1.23 seconds, the 4-bit model in 1.20 seconds. Time was significantly higher than the full precision model but that is to be expected as neuromorphic hardware is being simulated whereas the full precision model does not have this constraint. Going forward, we hope to tune the BrainChip model to increase its accuracy and time its operation in hardware, not just CPU simulation."



Loihi was done on a hardware cloud-based processor, while Akida was the MetaTF Akida simulation software, so the loihi time is distorted by the internet delay, while the Akida time derives from a software implementation.

 

[AusEire]

No issues promoting Brainchip in a respectful way. There are a few posters who do it brilliantly too.

Agree with not being arrogant to other company personnel, especially when Intel is our partner. It's a great opening for Brainchip to go next level.
US government is supporting Intel heavily for local businesses initiative, let's not kill the golden goose.

Thanks mate

I do my best

 

[BaconLover]

Thanks mate

I do my best

I'm sorry that's your best

Keep on trying, you'll make it one day ❤

 

[Mt09]

Intel’s reply lol.

Hello, @DavidMorris



Thank you for posting on the Intel® communities.
For more updates and information about Loihi 2, we recommend you to visit our Intel Newsroom and stay tuned with this portal: Intel Advances Neuromorphic with Loihi 2, New Lava Software Framework and New Partners.

Regarding Akida's subject, since this product was not designed by Intel, we recommend you to contact Akida's Developer for more inquiries.

Best regards,

Jocelyn M.

Intel Customer Support Technician.

 

[Rise from the ashes]

Intel’s reply lol.

Hello, @DavidMorris



Thank you for posting on the Intel® communities.
For more updates and information about Loihi 2, we recommend you to visit our Intel Newsroom and stay tuned with this portal: Intel Advances Neuromorphic with Loihi 2, New Lava Software Framework and New Partners.

Regarding Akida's subject, since this product was not designed by Intel, we recommend you to contact Akida's Developer for more inquiries.

Best regards,

Jocelyn M.

Intel Customer Support Technician.

At least that was a professional reply.

 

[Getupthere]

Top 17 ETFs with AU:BRN / BrainChip Holdings Ltd (ASX)

Top 17 ETFs with AU:BRN / BrainChip Holdings Ltd (ASX)

fintel.io

 

[TheDrooben]

Hi Larry,

I don't think the times really tell us much.

The test was done using the Akida ADE simulation software:

"3.3. Intel Loihi and BrainChip Akida
The 10-layer deep learning network was converted into formats compatible with Intel and BrainChip. In both cases, precision of the weights was varied to see how it affected model size and accuracy. Timing results were also produced.
For Intel, we accessed Loihi (Nahuku32) chips via the cloud and NxSDK 1.0;
for BrainChip, we simulated the chips on CPU via their hardware abstraction layer. Timing results are not comparable because CPU or GPU simulations are much faster than running on neuromorphic hardware."

5.
"Due to this constraint, our experiments were conducted using Intel’s cloud environment and the stand-alone simulation environment for Akida. There are drastic differences between software simulation and hardware and we intend to study those further."

However, this does explain something else. I saw a quote a couple of weeks ago which referred to 8-bit akida, and, as we all know, Akida only has 4 bits. However, apparently when using the Akida simulator, you can choose 8 bits.

"For BrainChip, accuracy also increased with precision; maximum accuracy was 82.5% at 8 bits and 66.8% at 4 bits. The 8-bit model produced results in 1.23 seconds, the 4-bit model in 1.20 seconds. Time was significantly higher than the full precision model but that is to be expected as neuromorphic hardware is being simulated whereas the full precision model does not have this constraint. Going forward, we hope to tune the BrainChip model to increase its accuracy and time its operation in hardware, not just CPU simulation."



Loihi was done on a hardware cloud-based processor, while Akida was the MetaTF Akida simulation software, so the loihi time is distorted by the internet delay, while the Akida time derives from a software implementation.

Thanks Dio.......glad to be corrected by the best. Send me the Ogre!!

Larry

 

[TopCat]

Here’s something to think about after some posts yesterday. IniVation CEO ,Kynen Eng, talks about their new tech. This was from yesterday. There’s no mention of SynSense in this article but instead refers to a new technology and working with different AI technologies. Is it remotely possible iniVation has followed Prophesee and now using Brainchip?

IniVation new ultra-high speed camera sensors make autonomous driving safer, says CEO Kynan Eng

IniVation, a Switzerland-based vision sensor and software company, introduced its latest high-performance neuromorphic vision sensor technology, Aeveon, at CES 2023. While the biology-inspired vision technology has many applications in highly specialized niche fields, particularly industrial...

www.digitimes.com

Compared to IniVation's previous technology: Dynamic Vision Sensor (DVS) which could only normally output pixel-level change events, Aeveon, using its new Adaptive Event Vision technology, is able to combine ultra-high speed event-based sensing to perform high HDR frame-based output and on-sensor compression that reduces output bandwidth requirements.

For autonomous vehicles and robotics, the company is working with different AI technologies that are compatible with the sensors. Eng said IniVation's mission is to make sure that the AI system receives data from the new sensor as fast as possible,