BRN Discussion Ongoing

[TasTroy77]

 

[VictorG]

New neuromorphic chip for AI on the edge, at a small fraction of the energy and size of today's computing platforms

An international team of researchers has designed and built a chip that runs computations directly in memory and can run a wide variety of AI applications–all at a fraction of the energy consumed by computing platforms for general-purpose AI computing.

techxplore.com

This looks like genuine competition

I'm no tech head but I cant see NeuRRam being in the same century as Akida much less the same market. my understanding is it uses convolution processing (analogue to digital). they are yet to tackle spiking architecture and their power saving is like 30 to 40 times less efficient than Akida. Furthermore the chip achieved 87% accuracy on image classification - imagine using it in self driving cars, it would make Tesla look good.

 

[Dozzaman1977]

New neuromorphic chip for AI on the edge, at a small fraction of the energy and size of today's computing platforms

An international team of researchers has designed and built a chip that runs computations directly in memory and can run a wide variety of AI applications–all at a fraction of the energy consumed by computing platforms for general-purpose AI computing.

techxplore.com

This looks like genuine competition

 

[uiux]

New neuromorphic chip for AI on the edge, at a small fraction of the energy and size of today's computing platforms

An international team of researchers has designed and built a chip that runs computations directly in memory and can run a wide variety of AI applications–all at a fraction of the energy consumed by computing platforms for general-purpose AI computing.

techxplore.com

This looks like genuine competition

Looks like a good chip

Gert Cauwenberghs was a member of the scientific advisory board at one stage and had positive things to say about BrainChip technology.

https://s3.amazonaws.com/content.stockpr.com/brainchipinc/db/217/1301/pdf/20160429+BRN+Virtual+Roadshow+Presentation+Release+ASX+Release.pdf

 

[Slade]

Looks like a good chip

Gert Cauwenberghs was a member of the scientific advisory board at one stage and had positive things to say about BrainChip technology.

https://s3.amazonaws.com/content.stockpr.com/brainchipinc/db/217/1301/pdf/20160429+BRN+Virtual+Roadshow+Presentation+Release+ASX+Release.pdf

View attachment 14420

Any breach of our patents I wonder.

 

[uiux]

Any breach of our patents I wonder.

Doubtful

Gert is another genius

 

[uiux]

There exists a video of Jeff Krichmar, Gert Cauwenbergs and Nicholas Spitzer talking up Brainchip



Maybe the really long long termers remember it and can source it?

 

[Slade]

There exists a video of Jeff Krichmar, Gert Cauwenbergs and Nicholas Spitzer talking up Brainchip



Maybe the really long long termers remember it and can source it?

They compliment BrainChip while secretly working on developing their own neuromorphic chip to compete with Akida. Have I got this right?

 

[uiux]

They compliment BrainChip while secretly working on developing their own neuromorphic chip to compete with Akida. Have I got this right?

Secretly? No

You can read Gerts published works and patents publicly for most of his career


Maybe you don't understand that they were on the scientific advisory board BECAUSE they are experts in the field

 

[Rskiff]

There exists a video of Jeff Krichmar, Gert Cauwenbergs and Nicholas Spitzer talking up Brainchip



Maybe the really long long termers remember it and can source it?

This one?

 

[uiux]

This one?

That's Toni Delbruck from inilabs

 

[Slade]

Secretly? No

You can read Gerts published works and patents publicly for most of his career


Maybe you don't understand that they were on the scientific advisory board BECAUSE they are experts in the field

Funny because today is the first day that I have read about the development of the NeuRRAM chip. When did you first hear about it?

What do you mean by maybe I don’t understand that they were on the scientific advisory board BECAUSE they are experts in the field?

 

[uiux]

Funny because today is the first day that I have read about the development of the NeuRRAM chip. When did you first hear about it?

What do you mean by maybe I don’t understand that they were on the scientific advisory board BECAUSE they are experts in the field?

I've been following Gerts work for 5 years or so, sure the chip is news, but the underlying innovations have been trickling out for ages

Gert Cauwenberghs

Professor of Bioengineering and Co-Director, Institute for Neural Computation, UC San Diego - Cited by 19,784 - Biomedical integrated circuits - neuromorphic systems engineering - learning and intelligent systems

scholar.google.com

 

[Slade]

I've been following Gerts work for 5 years or so, sure the chip is news, but the underlying innovations have been trickling out for ages

Gert Cauwenberghs

Professor of Bioengineering and Co-Director, Institute for Neural Computation, UC San Diego - Cited by 19,784 - Biomedical integrated circuits - neuromorphic systems engineering - learning and intelligent systems

scholar.google.com

So the chip was kept a secret. Interesting.

 

[uiux]

So the chip was kept a secret. Interesting.

Sure if that's how you want to internalise it

 

[VictorG]

Am I missing something here!!
Why is NueRram a good chip and how does it compare to Akida?

 

[Slade]

Sure if that's how you want to internalise it

Maybe you don’t understand that in order to gain market advantage, companies developing new technology like to keep things SECRETIVE.

 

[uiux]

Maybe you don’t understand that in order to gain market advantage, companies developing new technology like to keep things secretive.

Maybe I understand it and think the notion of Peter van der Made being so careless as to let his IP slip away from him is ridiculous

Further, the patent for this "new" chip is listed in Gerts scholar profile from 2021, so it's not "new" or "secret"

Secondly, the innovations seem radically different - you are assuming stolen IP when the two innovations aren't close

 

[Slade]

Maybe I understand it and think the notion of Peter van der Made being so careless as to let his IP slip away from him is ridiculous

Further, the patent for this "new" chip is listed in Gerts scholar profile from 2021, so it's not "new" or "secret"

Secondly, the innovations seem radically different - you are assuming stolen IP when the two innovations aren't close

Please don’t put words in my mouth. I asked a question and never made a statement. If you don’t know the answer that’s fine. Today is the first we have heard of this chip that to many of us sounds very similar to Akida. You tell us that one of the developers was on BrainChip’s scientific board. The question that I asked is reasonable. The guy goes from our scientific board and later is part of a team that releases a neuromorphic chip. Unlike Akida, which we knew about for years before it was developed, this chip seems to have come out of the blue. With all your research you didn’t even know it was being developed.

 

[uiux]

Please don’t put words in my mouth. I asked a question and never made a statement. If you don’t know the answer that’s fine. Today is the first we have heard of this chip that to many of us sounds very similar to Akida. You tell us that one of the developers was on BrainChip’s scientific board. The question that I asked is reasonable. The guy goes from our scientific board and later is part of a team that releases a neuromorphic chip. Unlike Akida, which we knew about for years before it was developed, this chip seems to have come out of the blue. With all your research you didn’t even know it was being developed.

I check Gerts profile every few weeks, so I would have seen the patent before



So what exactly are you saying then Slade?