BRN Discussion Ongoing

[Deadpool]

@Fact Finder has been employed by Brainchip as acting Legal Counsel.

Obviously he’s no longer able to make comments or express opinions in relations to the said company Brainchip Holdings .

He’s permitted to read what’s written on this forum & it’s my wish that if he chooses to read this post that he does NOT seek damages due to myself @Tothemoon24 speaking absolute crap as usual.

Funny you should say that , I heard a rumor that he was poached by OpenAI to instruct the chatgpt algo team, how it's done.

 

[Xray1]

I thought I'd read that PVDM wouldn't be attending the AGM somewhere here also, and had it attributed in my mind as coming from you Tech.
But just had a quick look back and cant find it now. Also recalled that the reason was that they were too busy at the moment to attend and recall whoever said it commenting on their sterling work ethic.
Weird because I remember thinking at the time that it's probably a relief for him so he won't have to answer about AI taking over the world a hundred times again. My dumb question.
I would be happy to see and thank both him and Anil if they wished to attend but understand the Company has grown beyond that necessity now.
In the early days everyone naturally wants to see the originator and get a feel for his credibility and ethics, but now that Brainchip has grown and employs professional management as not only guidance but also representation, it's probably not necessary.
The dealing with investors thing is all a part of Sean and the other business managers schtick, but think it probably something of a strain for our more tech oriented fellows who would have to be constantly watching their P's and Q's so as not to let something inappropriate slip in conversation with our wily dot joiners.

Agreed ....... I also thought that I'd read a post a while ago that PVDM wouldn't be attending the AGM somewhere here also but I can't locate the post here anymore.

 

[Boab]

Tim Llewellynn1.CEO/Co-Founder of NVISO Human Behaviour AI | President Bonseyes Community Association | Coordinator Bonseyes AI Marketplace | IBM Beacon Award Winner #metaverse #edgeai #decentralizedai
33 Min. •

And we haven't even arrived at #neuromorphic computing - there is so much performance improvement possible with Edge AI that applications in the next 5 years are going to be mind blowing. It will feel like Edge AI applications arrived from nowhere - but have effectively been over a decade in the making.
View attachment 34140

The interesting thing is it does have 50 times better performance per watt. I'm not a techy so I'm not sure how this may affect us?.
I think Tim is saying ok this is getting better but wait til we add some Akida??
Any thoughts ?

 

[DollazAndSense]

The interesting thing is it does have 50 times better performance per watt. I'm not a techy so I'm not sure how this may affect us?.
I think Tim is saying ok this is getting better but wait til we add some Akida??
Any thoughts ?

View attachment 34149

I take comfort that Rob Telson doesn’t seem to view NVIDIA as a competitor but as a partner in the future (as per some interview quite some time ago).

 

[jtardif999]

Well akida needs to be shown something then it gets labeled as per a video some time ago. Then it can look at something very similar and identify it .

That video with the elephant, police car and other things just sounded similar to what was explained in that article.



Been a long day for me so you may be right

Exactly the point I’m making - it is shown a picture of an elephant then it knows when shown an image of a figurine of an elephant or an image of a real elephant that they are all the same thing. It doesn’t need to be taught to learn this way, it just does so naturally. So when I saw the description about teaching AI to learn in a few shots I know it’s referring to AI generated by computers… and in fact when I read the article O found it was mostly referring to generative AI - the current flavour of the moment.

 

[TheDrooben]

Our university partners Carnegie Mellon win AFRL challenge. We have had collaboration with AFRL in the past through ISL

Carnegie Mellon University experts win Air Force Research Laboratory AI Grand Challenge

A team of collaborators from Carnegie Mellon University and the University of North Carolina at Chapel Hill won Phase 1 of the National Security Innovation Network (NSIN)-sponsored Air Force Research Laboratory (AFRL) Active AI Planners for Chemistry

indiaeducationdiary.in

BrainChip-Made AI Processor Used in New Tech for AFRL Radar Projects - ExecutiveBiz

BrainChip-Made AI Processor Used in New Tech for AFRL Radar Projects

blog.executivebiz.com

Larry

 

[TechGirl]

Haven't seen this before, it's on Brainchip's blog page, I haven't read it yet, once you click on link scroll down & click read full article.

The advantages of Unified Deep Learning technology

Explore valuable insights into this cutting-edge approach and discover how it revolutionizes the field of AI.

brainchip.com

The advantages of Unified Deep Learning technology​

The growth of Artificial Intelligence (AI) has been rapid, with the market heavily focused on training large generative models, such as Open AI’s chatGPT. Deep learning models currently rely heavily on neural networks (NNs), which require high-power, massively parallel computing to calculate the millions of values needed for each inference instance. This results in the cost of training a model of such a large caliber requiring millions of dollars in computational power.
This paper explores an alternative, neuromorphic computing, which can be up to 1,000 times better performance and 10,000 times better efficiency compared to traditional high-performance computing hardware, such as CPUs and GPUs. Neuromorphic computing also reduces the need for high-power cloud inference, raw data traffic, and congestion in networks. However, there are very few Spiking Neural Network (SNN) models that are ready for use on the market.
BrainChip has developed the Akida technology, which combines convolution functions efficiently with
a fully digital, neuromorphic computing core. The Akida technology is capable of executing most deep learning networks and performing inference at an energy cost that is a fraction of conventional solutions, such as convolutional neural networks (CNNs) and deep neural networks (DNNs). The radical energy efficiency of neuromorphic computing opens the capability of learning in real-time and in the field, enabling immediate customization of AI-enhanced products.
Click here to read the full article.

This is link to full article not sure if it will work?

https://brainchip.com/wp-content/uploads/2023/03/The_advantages_of_Unified_Deep_Learning_technology.pdf

 

[Boab]

Haven't seen this before, it's on Brainchip's blog page, I haven't read it yet, once you click on link scroll down & click read full article.

The advantages of Unified Deep Learning technology

Explore valuable insights into this cutting-edge approach and discover how it revolutionizes the field of AI.

brainchip.com

The advantages of Unified Deep Learning technology​

The growth of Artificial Intelligence (AI) has been rapid, with the market heavily focused on training large generative models, such as Open AI’s chatGPT. Deep learning models currently rely heavily on neural networks (NNs), which require high-power, massively parallel computing to calculate the millions of values needed for each inference instance. This results in the cost of training a model of such a large caliber requiring millions of dollars in computational power.
This paper explores an alternative, neuromorphic computing, which can be up to 1,000 times better performance and 10,000 times better efficiency compared to traditional high-performance computing hardware, such as CPUs and GPUs. Neuromorphic computing also reduces the need for high-power cloud inference, raw data traffic, and congestion in networks. However, there are very few Spiking Neural Network (SNN) models that are ready for use on the market.
BrainChip has developed the Akida technology, which combines convolution functions efficiently with
a fully digital, neuromorphic computing core. The Akida technology is capable of executing most deep learning networks and performing inference at an energy cost that is a fraction of conventional solutions, such as convolutional neural networks (CNNs) and deep neural networks (DNNs). The radical energy efficiency of neuromorphic computing opens the capability of learning in real-time and in the field, enabling immediate customization of AI-enhanced products.
Click here to read the full article.

This is link to full article not sure if it will work?

https://brainchip.com/wp-content/uploads/2023/03/The_advantages_of_Unified_Deep_Learning_technology.pdf

After reading that I want to buy more shares but there are no funds available.
Thanks for sharing

 

[Deena]

According to Morningstar Quantitative Ratings, Brainchip is now an astonishing 62.84% undervalued with fair value at $1.238.
Just imagine what it will be once the next big news hits, or the shortly to be released quarterly shows an improving revenue stream!
Deena

 

[Bravo]

What about Jetson Orin compared to Akida?

Hi @Dhm, Here's something from the NVISO November 2022 presentation. Maybe they don't want to make it too easy for their competitors to join the dots.

 

[Bravo]

When comparing chalk and cheese, are we talking camembert or parmesan?

Perceive's figures are for the YOLO V5 database.

In 2019, the Akida simulator did 30 fps @ 157 mW on MobileNet V1.

View attachment 27636
I don't have the fps/W figures for the Akida 1 SoC performance, so if anyone has those to hand, it would be much appreciated.

The chip simulator seems to have topped out at 80 fps. We know Akida 1 SoC can top 1600 fps (don't recall what database), which is 20 times faster than the simulator, so it wouldn't even get into second gear doing 30 fps.

Back in 2018, Bob Beachler said they expected 1400 frames per second per Watt.
https://www.eejournal.com/article/brainchip-debuts-neuromorphic-chip/

So 30 fps would be 20 mW if there is a linear dependence. And, of course, we were told that the SoC performed better than expected.

But the main thing is that the comparison databases need to be the same as performance varies depending on database.

Perceive relys on compression to achieve extreme sparsity by ignoring the zeros in the multiplier.

US11003736B2 Reduced dot product computation circuit

View attachment 27638

[0003] Some embodiments provide an integrated circuit (IC) for implementing a machine-trained network (e.g., a neural network) that computes dot products of input values and corresponding weight values (among other operations). The IC of some embodiments includes a neural network computation fabric with numerous dot product computation circuits in order to process partial dot products in parallel (e.g., for computing the output of a node of the machine-trained network). In some embodiments, the weight values for each layer of the network are ternary values (e.g., each weight is either zero, a positive value, or the negation of the positive value), with at least a fixed percentage (e.g., 75%) of the weight values being zero. As such, some embodiments reduce the size of the dot product computation circuits by mapping each of a first number (e.g., 144) input values to a second number (e.g., 36) of dot product inputs, such that each dot product input only receives at most one input value with a non-zero corresponding weight value.

1. A method for implementing a machine-trained network that comprises a plurality of processing nodes, the method comprising:

at a particular dot product circuit performing a dot product computation for a particular node of the machine-trained network:

receiving (i) a first plurality of input values that are output values of a set of previous nodes of the machine-trained network and (ii) a set of machine-trained weight values associated with a set of the input values;

selecting, from the first plurality, a second plurality of input values that is a smaller subset of the first plurality of input values, said selecting comprising (i) selecting the input values from the first plurality of input values that are associated with non-zero weight values, and (ii) not selecting a group of input values from the first plurality of input values that are associated with weight values that are equal to zero;

computing a dot product based on (i) the second plurality of input values and (ii) weight values associated with the second plurality of input values.

ASIDE: This isn't relevant to the discussion, but I stumbled across it just now and it shows that we developed a dataset for MagikEye in 2020, which I had not observed before.
View attachment 27637

@Diogenese "I don't have the fps/W figures for the Akida 1 SoC performance, so if anyone has those to hand, it would be much appreciated".

Here we go Dodgy Knees! Better late than never I always say!

The Linley Group Microprocessor Report - October 2019

https://d1io3yog0oux5.cloudfront.net/brainchipinc/files/BrainChip+Akida+Is+a+Fast+Learner.pdf

 

[wasMADX]

@Fact Finder has been employed by Brainchip as acting Legal Counsel.

Obviously he’s no longer able to make comments or express opinions in relations to the said company Brainchip Holdings .

He’s permitted to read what’s written on this forum & it’s my wish that if he chooses to read this post that he does NOT seek damages due to myself @Tothemoon24 speaking absolute crap as usual.

Due to "digital deluge" I have limited my time here recently & am out of touch. Also, somehow I have not been receiving BRN newsletters lately to which I subscribed years ago.

How did you find out about FF as our Legal Counsel please? I obviously need to change some settings here.

I'm thrilled that FF has been employed by us. If that happening had been announced on the ASX, we would/should have seen it reflected in a SP increase. Who better than FF to filter fact from fiction on this forum and keep our company up to speed with our forum findings? He won't be limiting himself to stuffy legal matters.

 

[Deleted member 118]

Due to "digital deluge" I have limited my time here recently & am out of touch. Also, somehow I have not been receiving BRN newsletters lately to which I subscribed years ago.

How did you find out about FF as our Legal Counsel please? I obviously need to change some settings here.

I'm thrilled that FF has been employed by us. If that happening had been announced on the ASX, we would/should have seen it reflected in a SP increase. Who better than FF to filter fact from fiction on this forum and keep our company up to speed with our forum findings? He won't be limiting himself to stuffy legal matters.

 

[ndefries]

@Fact Finder has been employed by Brainchip as acting Legal Counsel.

Obviously he’s no longer able to make comments or express opinions in relations to the said company Brainchip Holdings .

He’s permitted to read what’s written on this forum & it’s my wish that if he chooses to read this post that he does NOT seek damages due to myself @Tothemoon24 speaking absolute crap as usual.

Is this true? How did you find out?

 

[Deleted member 118]

Is this true? How did you find out?

 

[Boab]

View attachment 34169

In English please

 

[Deleted member 118]

Just in case anyone else asks

 

[Boab]

Just in case anyone else asks

Me neither

 

[wasMADX]

Sony investment will put AI chips inside Raspberry Pi boards​

Steve Dent
April 12, 2023, 6:35 pm

Sony's semiconductor division has announced that it's making a "strategic investment" in Raspberry Pi as a way to bring its AI tech to a wider market. The idea is to give Raspberry Pi users around the world a development platform for its Aitrios edge computing (on-chip) AI platform used for image sensing functions like facial recognition.
"We are very pleased to be partnering with Raspberry Pi Ltd. to bring our Aitrios platform — which supports the development of unique and diverse solutions utilizing our edge AI devices — to the Raspberry Pi user and developer community, and provide a unique development experience," said Sony Semiconductor Solutions president and CEO Terushi Shimizu.

The Raspberry Pi 4 and other devices from the company give users PC-like power in a small form factor. Originally designed as an educational platform to teach robotics, coding and more, it has become popular as a way for coders to prototype IoT (Internet of Things) and other devices.
The addition of Sony's Aitrios could make it even more useful. Unlike cloud AI, it runs directly on chips (edge computing) to reduce latency, and Sony has pitched the system for uses like surveillance, security and more. Examples cited on a dedicated website include inventory monitoring and retention, customer counting, license plate recognition and "detailed employee analysis." Sony says it preserves privacy by analyzing data strictly on-chip and only sending metadata to the cloud.
Sony is already involved with Raspberry Pi as a "longstanding and valued strategic partner," the company said. It recently provided imaging chips with autofocus capability and helped Raspberry Pi get its UK manufacturing plant up to speed in the early days of the company.

Isn't this a kick in the guts/missed opportunity for us? Years ago I contacted electronics hobbyist magazines to ensure they knew about Akida, but I didn't think to notify the Raspberry Pi and Arduino people. Is it too late to incorporate us into those platforms?

 

[Bravo]

Due to "digital deluge" I have limited my time here recently & am out of touch. Also, somehow I have not been receiving BRN newsletters lately to which I subscribed years ago.

How did you find out about FF as our Legal Counsel please? I obviously need to change some settings here.

I'm thrilled that FF has been employed by us. If that happening had been announced on the ASX, we would/should have seen it reflected in a SP increase. Who better than FF to filter fact from fiction on this forum and keep our company up to speed with our forum findings? He won't be limiting himself to stuffy legal matters.

Is this true? How did you find out?

I think someone is pulling your legs.

Our resident super sleuth and celebrity brainiac is taking a wee hiatus. Hopefully he doesn't take too loooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooong to return!