BRN Discussion Ongoing

[Fact Finder]

@Diogenese I literally have no idea what this means but is is good for BrainChip?

View attachment 19749

François Piednoël de Normandie on LinkedIn: #idc | 10 comments

Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn

www.linkedin.com

I think the following linked article is probably the starting point for François's post:

Ten Year Anniversary of Core 2 Duo and Conroe: Moore’s Law is Dead, Long Live Moore’s Law

www.anandtech.com

My opinion only DYOR
FF

AKIDA BALLISTA
PS: Definitely worth a read it reinforces why being like Fonzie is best:

 

[Diogenese]

@Diogenese I literally have no idea what this means but is is good for BrainChip?

View attachment 19749

François Piednoël de Normandie on LinkedIn: #idc | 10 comments

Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn

www.linkedin.com

Hi Tls,

Transistors have internal resistance, so, when a current flows, heat is generated. If we assume that each operation of a digital transistor switch (CMOS) in any specified geometric chip size uses the same amount of power no matter what chip it is in, then more TOPS means more power. Increasing the clock speed increases power consumption. In an IC, heat dissipates more slowly than it is generated, so there is an accumulation of heat. So to reduce power while obtaining an equivalent outcome means that the number of transistor switch operations must be reduced.

Pumping cooling fluid around and blowing air from a fan through a radiator much like a car's cooling system uses more power.

There are a couple of ways of reducing the number of switch operations, eg, not implementing multiply by zero operations in MAC, or lowering the voltage supply (which has the effect of reducing the signal-to-noise ratio (increasing error-proneness)).

However, these savings are small beer compared to the use of spikes instead of 8-bit or upward bytes to implement a function.

The article is talking about an Intel CPU, not Loihi, so it would be using 64-bit MACs or higher - very high precision, but very power-hungry.

Before Akida, most hardware SNNs used analog ReRAM/MemRistor because it is more closely analogous to real neurons. However, human technology lacks the millions of years of evolutionary refinement which lead to the human brain, and the manufacture of analog neurons lacks the precision necessary to derive consistently accurate outcomes. The performance of analog neurons is too variable, and considering that millions of neurons are included in an SNN, the accumulated errors can be significant. While some attempts to overcome this problem have been made, they generally end up using more power.

As we've discussed before, the discovery by Simon Thorpe's group that the receptor cells in the eye respond more quickly to stronger input signals, so most of the visual information is included in the earlier receptor cell signals. This can also be applied to the pixels of a camera's light sensor.

This lead to N-of-M coding in which only, say, 10% of incoming pixel data needs to be processed by choosing the first 10% of incoming spikes.

A lot of work of a CPU or GPU is image-related in that, apart from any image/video processing, the processor must control the screen display. I wonder if there is a role for Akida there?

 

[thelittleshort]

Hi Tls,

Transistors have internal resistance, so, when a current flows, heat is generated. If we assume that each operation of a digital transistor switch (CMOS) in any specified geometric chip size uses the same amount of power no matter what chip it is in, then more TOPS means more power. Increasing the clock speed increases power consumption. In an IC, heat dissipates more slowly than it is generated, so there is an accumulation of heat. So to reduce power while obtaining an equivalent outcome means that the number of transistor switch operations must be reduced.

Pumping cooling fluid around and blowing air from a fan through a radiator much like a car's cooling system uses more power.

There are a couple of ways of reducing the number of switch operations, eg, not implementing multiply by zero operations in MAC, or lowering the voltage supply (which has the effect of reducing the signal-to-noise ratio (increasing error-proneness)).

However, these savings are small beer compared to the use of spikes instead of 8-bit or upward bytes to implement a function.

The article is talking about an Intel CPU, not Loihi, so it would be using 64-bit MACs or higher - very high precision, but very power-hungry.

Before Akida, most hardware SNNs used analog ReRAM/MemRistor because it is more closely analogous to real neurons. However, human technology lacks the millions of years of evolutionary refinement which lead to the human brain, and the manufacture of analog neurons lacks the precision necessary to derive consistently accurate outcomes. The performance of analog neurons is too variable, and considering that millions of neurons are included in an SNN, the accumulated errors can be significant. While some attempts to overcome this problem have been made, they generally end up using more power.

As we've discussed before, the discovery by Simon Thorpe's group that the receptor cells in the eye respond more quickly to stronger input signals, so most of the visual information is included in the earlier receptor cell signals. This can also be applied to the pixels of a camera's light sensor.

This lead to N-of-M coding in which only, say, 10% of incoming pixel data needs to be processed by choosing the first 10% of incoming spikes.

A lot of work of a CPU or GPU is image-related in that, apart from any image/video processing, the processor must control the screen display. I wonder if there is a role for Akida there?

Thanks for taking the time to explain @Diogenese you legend

 

[thelittleshort]

AKIDA BALLISTA
PS: Definitely worth a read it reinforces why being like Fonzie is best:

View attachment 19750

 

[Izzzzzzzzzzy]

@Diogenese I literally have no idea what this means but is is good for BrainChip?

View attachment 19749

François Piednoël de Normandie on LinkedIn: #idc | 10 comments

Intel is now in position it was just before the launch of Conroe. Because competitive pressure, it was forced to push the power envelope and increase… | 10 comments on LinkedIn

www.linkedin.com

Talking about processors nothing to do with akida or intels Lohi

 

[jtardif999]

Hi Tls,

Transistors have internal resistance, so, when a current flows, heat is generated. If we assume that each operation of a digital transistor switch (CMOS) in any specified geometric chip size uses the same amount of power no matter what chip it is in, then more TOPS means more power. Increasing the clock speed increases power consumption. In an IC, heat dissipates more slowly than it is generated, so there is an accumulation of heat. So to reduce power while obtaining an equivalent outcome means that the number of transistor switch operations must be reduced.

Pumping cooling fluid around and blowing air from a fan through a radiator much like a car's cooling system uses more power.

There are a couple of ways of reducing the number of switch operations, eg, not implementing multiply by zero operations in MAC, or lowering the voltage supply (which has the effect of reducing the signal-to-noise ratio (increasing error-proneness)).

However, these savings are small beer compared to the use of spikes instead of 8-bit or upward bytes to implement a function.

The article is talking about an Intel CPU, not Loihi, so it would be using 64-bit MACs or higher - very high precision, but very power-hungry.

Before Akida, most hardware SNNs used analog ReRAM/MemRistor because it is more closely analogous to real neurons. However, human technology lacks the millions of years of evolutionary refinement which lead to the human brain, and the manufacture of analog neurons lacks the precision necessary to derive consistently accurate outcomes. The performance of analog neurons is too variable, and considering that millions of neurons are included in an SNN, the accumulated errors can be significant. While some attempts to overcome this problem have been made, they generally end up using more power.

As we've discussed before, the discovery by Simon Thorpe's group that the receptor cells in the eye respond more quickly to stronger input signals, so most of the visual information is included in the earlier receptor cell signals. This can also be applied to the pixels of a camera's light sensor.

This lead to N-of-M coding in which only, say, 10% of incoming pixel data needs to be processed by choosing the first 10% of incoming spikes.

A lot of work of a CPU or GPU is image-related in that, apart from any image/video processing, the processor must control the screen display. I wonder if there is a role for Akida there?

“Before Akida, most hardware SNNs used analog ReRAM/MemRistor because it is more closely analogous to real neurons.”

Hi @Diogenese, great post I have been pondering Qualcomm and their analog SNN patents for a while. I remember when BRN used to mark them as a competitor by listing them along with Intel and IBM as tech comparisons - you know the checklist ticks against the different categories such as continuous learning and power consumption etc. Then they stopped including them in that list around 2018 or 19. I assumed when that happen that perhaps they’ve bought the carrot and would be a future customer. There’s been a lot said about the inherent instability in analog SNNs. Do we think that somehow Qualcomm have solved this? Not likely, unless they’ve done so somewhere in the last 18 months and patents will soon be published to reflect that. That would be a pretty big deal considering academia world wide has yet to bare such an accomplishment. Not to say it’s impossible for Qualcomm but you would think they would need to have harnessed the crème of Nueromorphic talent from academia to do so. So perhaps it’s then possible that Qualcomm’s Snapdragon platform includes a bit of our magic to make it all work, I think so anyway, just rambling on a bit haven’t done so for a while - thanks to all the great contributors to this forum I haven’t felt the need to. Cheers to all.

 

[Fact Finder]

I found the following research article and was initially attracted because I thought lip reading that would be of use to the hearing impaired and of course it would be but being a Chinese sponsored and Chinese located research project the aim is for mass not individual lip reading.

Every tyrants dream the ability to read the lips of every person in a public place or street.

Pingo cannot control minds just yet but he could eventually control lips which is one step further than simply controlling speech:

https://www.researchgate.net/publication/364395637_Lipreading_Using_Liquid_State_Machine_with_STDP-Tuning

What science offers the world seldom scares me but this is going many, many steps too far for mine.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[TopCat]

I found the following research article and was initially attracted because I thought lip reading that would be of use to the hearing impaired and of course it would be but being a Chinese sponsored and Chinese located research project the aim is for mass not individual lip reading.

Every tyrants dream the ability to read the lips of every person in a public place or street.

Pingo cannot control minds just yet but he could eventually control lips which is one step further than simply controlling speech:

https://www.researchgate.net/publication/364395637_Lipreading_Using_Liquid_State_Machine_with_STDP-Tuning

What science offers the world seldom scares me but this is going many, many steps too far for mine.

My opinion only DYOR
FF

AKIDA BALLISTA

Not quite sure where I’ve read it or if I’m even confused with something else , but I’m sure I’ve come across an article about this and it’s even very accurate if a mask is worn.

 

[TopCat]

New System Allows Lip-Reading Through Masks Using Wi-Fi and Radio Frequencies - The Debrief

Researchers have revealed a system that uses radio frequency (RF) and Wi-Fi signals for Lip-reading when the speaker is wearing a mask.

thedebrief.org

 

[Boab]

Previous 4C dates.
Wed 27th Jan 2021
Wed 28th April 2021
Thurs 29th July 2021
Thurs 21st Oct....along with 2 other announcements. 1) Taking orders of AKIDA AI
2) US Patent granted.
Thurs 27th Jan 2022
Tues 26th April 2022 with Monday being ANZAC day.
Wed 27th July.

If we go with the average it'll be Wed or Thurs for the upcoming 4C.

 

[Fact Finder]

New System Allows Lip-Reading Through Masks Using Wi-Fi and Radio Frequencies - The Debrief

Researchers have revealed a system that uses radio frequency (RF) and Wi-Fi signals for Lip-reading when the speaker is wearing a mask.

thedebrief.org

These researchers are trying to help the hearing impaired but if their technology advance was combined with this Chinese approach there would be no hiding.

 

[MDhere]

In 2018 according to Google there was over 64 million kilometres of paved roads in the world.

At 4 AKIDA chips to a kilometre that is a lot of chips Blu Wireless would need to cover the needs of automotive. Add in an AKIDA chip in every connected vehicle and that becomes a very large number???

My opinion only DYOR
FF

AKIDA BALLISTA

All this Blu talk has made yr black dancing poodle want to

change colour

 

[Fact Finder]

All this Blu talk has made yr black dancing poodle want to View attachment 19764 change colour

A Brainchip poodle:

 

[Deleted member 118]

Thanks for taking the time to explain @Diogenese you legend

I bet you still don’t understand

 

[Moneytalks]

Happy Sunday everyone Back in Akida Research HQ (Perth) after a great break to see family/friends in Canada.
Just wanted to share a quick experience that hopefully we can all soon relate to in the very near future. While enjoying the broadcast of the US Open and various NFL games I would use the ad breaks to catch up on the dot joining happening on TSE. I can't explain to you the pleasure of hearing the words "neuromorphic computing 5x more intelligent then current voice AI" come blaring through the TV screen and the smile it brought to my face. Mercedes are pushing their advertising hard. I got to hear those words and see that ad on multiple occasions.
We all know it's Akida they are talking about and we are witnessing it becoming part of the vernacular
Patience still needed and, like many here, not expecting too much in the quarterly this week........ But it is coming. No question. Here's to a green Monday!

Akida Ballista

 

[Iseki]

Good luck to all holders this week. I hope we have >$2 million revenue and a business as usual 4C

If not, these things cheer me up:

0. neuromorphic chips that can do something like autonomous adaptation are a hot hot hot topic and only akida can do it.
1. loihi and true north etc are not available in an Arm based microcontroller, or a SiFive one either
2. Arm is not going to be sold to NVIDIA who gave us the GPU's to simulate neuromorphic
3. Intel failed to buy SiFive.
4. There are many many AI chips coming out right now with Arm + floating calc units, building up a demand that akida based chips can take.

So, we are lucky, indeed. A demand is being created for us and we can't be blocked.

If I hope for anything more it might be $-Outgoings spent on development eg can intel's lava (should it not be larvae?) be ported to akida ip?

At any rate, the world is moving quickly, but so far it suits us.

 

[Deleted member 118]

Well I’m

 

[Tothemoon24]

Evening Chippers.
Have we discussed the Mercedes-Benz
“Attention Assist “currently being used in the EQS ….?


If you care to google Mercedes - Benz - Awake
a nice surprise awaits if not already mentioned

 

[equanimous]

Came across this, dont think it was posted before

Market Research Reports on Industry Analysis & Forecasts - Market Insights Reports

MarketInsightsReports.com provides market research reports and industry analysis on products, services, companies, markets.

www.marketinsightsreports.com

 

[equanimous]

ok so there is alot to unpack here which I quickly skimmed through.

Energy-Efficient Architecture and Dataflow Optimization for Spiking Neural Network Accelerators

Author(s): Lee, Jeong-Jun | Advisor(s): Li, Peng | Abstract: Spiking neural networks (SNNs) offer a promising biologically-plausible computing model and lend themselves to ultra-low-power event-driven processing on neuromorphic processors. Compared with the conventional artificial neural...

escholarship.org