BRN Discussion Ongoing

[Dallas]

European Space Agency - ESA on LinkedIn: Within the span of a few hours we released two innovative missions into…

Within the span of a few hours we released two innovative missions into space, taking a ride on a SpaceX Falcon 9 rocket. The Arctic Weather Satellite is set…

www.linkedin.com

 

[manny100]

Will have to catch up for a beer if it ever makes $1 again lol

IMO there is a fair chance that the patent portfolio alone is worth more than that.

 

[DingoBorat]

x.com

x.com

They have an idea, but yet to materialise it, by the looks of things..

Fractile

www.fractile.ai

Main guy is a young gun..

"Founded in 2022 by 28-year-old artificial intelligence PhD, Walter Goodwin, Fractile s building its first new AI chip, capable of running state-of-the-art AI models up to 100x faster and 10x cheaper than existing hardware"

 

[Dallas]

Frontgrade Gaisler on LinkedIn: Congratulations to the European Space Agency - ESA, Rymdstyrelsen (SNSA)…

Congratulations to the European Space Agency - ESA, Rymdstyrelsen (SNSA), OHB Sweden and its industrial team to the successful launch of the Arctic Weather…

www.linkedin.com

 

[Diogenese]

They have an idea, but yet to materialise it, by the looks of things..

Fractile

www.fractile.ai

Main guy is a young gun..

"Founded in 2022 by 28-year-old artificial intelligence PhD, Walter Goodwin, Fractile s building its first new AI chip, capable of running state-of-the-art AI models up to 100x faster and 10x cheaper than existing hardware"

GB2625821A Analog neural network = WO2024141751A1 ANALOG NEURAL NETWORK 20221229

GOODWIN WALTER THOMAS ROMBOLD [GB]

NEU EDGE LTD [GB]

An analog neural network is described comprising: a plurality of layers connected to form an electrical circuit having an input and an output, the input suitable for receiving an electrical signal corresponding to an input example and the output corresponding to an output of the neural network. Each layer comprises elements connected together, where the elements comprise: at least one programmable electronic element representing a weight of the neural network; at least one non-linear element; at least one amplifier block; an error element. Each layer also comprises a measurement element for measuring a change in an electrical signal across the error element.

 

[Diogenese]

GB2625821A Analog neural network = WO2024141751A1 ANALOG NEURAL NETWORK 20221229

GOODWIN WALTER THOMAS ROMBOLD [GB]

NEU EDGE LTD [GB]

View attachment 68143


An analog neural network is described comprising: a plurality of layers connected to form an electrical circuit having an input and an output, the input suitable for receiving an electrical signal corresponding to an input example and the output corresponding to an output of the neural network. Each layer comprises elements connected together, where the elements comprise: at least one programmable electronic element representing a weight of the neural network; at least one non-linear element; at least one amplifier block; an error element. Each layer also comprises a measurement element for measuring a change in an electrical signal across the error element.

Oh dear!

https://fortune.com/2024/07/26/fractile-ai-chip-startup-nvidia-15-million-funding-seed-round/

Startup with ‘radical’ concept for AI chips emerges from stealth with $15 million to try to challenge Nvidia​

BYJeremy Kahn
July 26, 2024 at 4:00 PM

...

Fractile was founded in 2022, but has been operating for two years in “stealth mode,” while working on its chip designs. The company has secured its investment seed round from Kindred Capital, the innovation fund of the defense alliance NATO, and Oxford Science Enterprises, which led the funding. Also participating are Cocoa and Innovia Capital, as well as prominent angel investors and alumni of AI and semiconductor companies.

...

AI startup Groq, which already has its chips in production and offers them through its own cloud-based AI computing service, is following a somewhat similar approach, moving the system’s memory closer to where the processing takes place. Groq uses SRAM (static random access memory) components co-located them on the chip, rather than off-chip DRAM to do this. But Goodwin says Fractile is going a step further and combining memory and processing into a single component, which should mean Fractile’s chips are even faster.

So far, though, Fractile has only tested its designs in computer simulations and has yet to manufacture test chips. But Goodwin said Fractile is convinced from these simulations that it can run a large language model, the kind of AI models that power today’s consumer chatbots and form the foundation of most generative AI applications, 100 times faster and 10 times cheaper than Nividia’s GPUs.

 

[Diogenese]

Very interesting listen, especially after 38.30 onwards.

Georges Massing’s official title has clearly escaped the attention of his company’s marketing team: “Vice President MBOS Automated Driving, Powernet & Integration E/E”. We like to say that he is the global leader of autonomy and cabin technologies (amongst a few other things) for Mercedes-Benz AG, a brand known for safety, reliability, and luxury. Georges is part of the generation of Mercedes managers who will translate those values into a future where vehicle electronics increasingly define the driver experience and sense of value.

“We have two visions that are driving us. One: vision zero, so safety, zero accidents. The second vision that is driving us is how can we give back time to customer?" – Georges Massing

Giving time back to the customer means automated driving, and in this regard, Georges’ team leads the world: S-class sedans are the only passenger vehicle in the world certified for Level 3 driving with a system the company calls “Drive Pilot.” This means the driver, while needing to remain available to take control, is free to engage in non-driving tasks while Drive Pilot is active. This ability to leave drivers out of the loop is a critical differentiator with all other systems.

“No matter how good an assisted system is, you as a driver are still responsible and have to control the whole vehicle. As soon as we move to an automated system, now you as a driver, you give control back to the car and you can do way much more.” – Georges Massing

To maintain the brand’s focus on safety, Georges emphasizes the need for redundancy in sensing and equipment packages on the vehicle. While pushing new technologies such as end-to-end neural networks, the company is highly sensitive to being able to formally verify the safety and adaptability of advances.

Currently, Drive Pilot limits itself to favorable conditions: clear lane markings on approved freeways, moderate to heavy traffic with speeds under 40 MPH, daytime lighting and clear weather, with the driver visible to a camera located above the driver's display, and no construction zone present. Clearly, the future is about relaxing these conditions, with a relaxation of the speed limit a likely event in the near future.


“Drive Pilot is just the beginning of what can be done." – Georges Massing

TT

Hi TT,

"end-to-end neural networks" bodes well for BRN - and redundancy means two of everything!

Mercedes has previously mentioned that it places emphasis on standardization, so that may mean they will use the same tech for all sensors.

So far as we know, they are not using Akida silicon yet, but are doing signal processing in software, but, if the software is Akida/TeNNs simulation software, it will still beat the socks of the competition.

 

[DingoBorat]

Oh dear!

https://fortune.com/2024/07/26/fractile-ai-chip-startup-nvidia-15-million-funding-seed-round/

Startup with ‘radical’ concept for AI chips emerges from stealth with $15 million to try to challenge Nvidia​

BYJeremy Kahn
July 26, 2024 at 4:00 PM

...

Fractile was founded in 2022, but has been operating for two years in “stealth mode,” while working on its chip designs. The company has secured its investment seed round from Kindred Capital, the innovation fund of the defense alliance NATO, and Oxford Science Enterprises, which led the funding. Also participating are Cocoa and Innovia Capital, as well as prominent angel investors and alumni of AI and semiconductor companies.

...

AI startup Groq, which already has its chips in production and offers them through its own cloud-based AI computing service, is following a somewhat similar approach, moving the system’s memory closer to where the processing takes place. Groq uses SRAM (static random access memory) components co-located them on the chip, rather than off-chip DRAM to do this. But Goodwin says Fractile is going a step further and combining memory and processing into a single component, which should mean Fractile’s chips are even faster.

So far, though, Fractile has only tested its designs in computer simulations and has yet to manufacture test chips. But Goodwin said Fractile is convinced from these simulations that it can run a large language model, the kind of AI models that power today’s consumer chatbots and form the foundation of most generative AI applications, 100 times faster and 10 times cheaper than Nividia’s GPUs.

Your "Oh dear!" Doesn't indicate a great deal of concern @Diogenese

Is it the fact they are playing with analog, basing results off of simulations of analog, or something else?

They seem a way off down the track, but could probably have their first Engineering Samples, within the next 6 months, to a year?..

How does TENNs compare, with what they are trying to do for LLMs, in your opinion?

 

[Diogenese]

Your "Oh dear!" Doesn't indicate a great deal of concern @Diogenese

Is it that fact they are playing with analog, basing results off of simulations of analog, or something else?

They seem a way off down the track, but could probably have their first Engineering Samples, within the next 6 months, to a year?..

How does TENNs compare, with what they are trying to do for LLMs, in your opinion?

Raising 15 million pounds on the suggestion that analog is a radical concept is a bit of a stretch.

Their patent talks about clamping the input and output signal levels of the neurons, which they do using digital-to-analog and analog-to-digital converters. They also use error-correction.

In addition, they have only run simulation software. I wonder how much this software takes account of actual silicon manufacturing variations.

Astonishingly the international patent search only found one relevant prior art document. The patent examiner must be living in the Tardus, or maybe they used ChatGPT.

While analog does have theoretical benefits in simplicity, the need for the ADC & DAC for each neuron militates against these advantages..

As to TeNNs, it really is a radical concept on top of the radical concept of digital SNNs, but a technical comparison without any Fractile benchmarking is beyond my pay grade.

 

[Pom down under]

Sure thing

View attachment 68135

They say, where there's smoke, there's fire and we're seeing plenty of smoke.

Unless it's smoke and mirrors ...

Glass of water it will be then

 

[DingoBorat]

Raising 15 million pounds on the suggestion that analog is a radical concept is a bit of a stretch.

Their patent talks about clamping the input and output signal levels of the neurons, which they do using digital-to-analog and analog-to-digital converters. They also use error-correction.

In addition, they have only run simulation software. I wonder how much this software takes account of actual silicon manufacturing variations.

Astonishingly the international patent search only found one relevant prior art document. The patent examiner must be living in the Tardus, or maybe they used ChatGPT.

While analog does have theoretical benefits in simplicity, the need for the ADC & DAC for each neuron militates against these advantages..

As to TeNNs, it really is a radical concept on top of the radical concept of digital SNNs, but a technical comparison without any Fractile benchmarking is beyond my pay grade.

"Astonishingly the international patent search only found one relevant prior art document. The patent examiner must be living in the Tardus, or maybe they used ChatGPT"

Yeah, I imagined them "beating a path" through a seemingly "undiscovered" area, only to find that path, had just been partially overgrown and wasn't actually new ground..

But figured, it was just my lack of understanding..

 

[mrgds]

hi @Diogenese .................. like to ask you if you would agree that SNR (signal to noise ratio) could be seen as the same principals as SNNs ?
And would the use of SNNs give a higher SNR?
Appreciate your thoughts.
Cheers

 

[Xray1]

1. AKD1000 SoC – A limited run of chips was produced with no plans for further production. The chips did not generate significant revenue or sell and are now being repurposed for applications like edge boxes.

2. AKD1000 IP – Two IP licenses have been sold (Renesas and Megaship), and Sean mentions that ongoing engagement continues with others.

3. AKD1500 Chip - The AKD1500 is an accelerator reference chip, which assists partners in developing and demonstrating their solutions as a stepping stone to integrating the Akida IP into their production SoCs - It’s not meant to be a revenue stream (just like edge boxes).

4. AKD1500 IP – No licenses have been sold. However, it's worth noting that Megachip said on LinkedIn they played a role in developing Akida1500, which I found interesting, as it suggested they might have had an end client or their own use in mind.

5. AKIDA 2.0 IP – No licenses sold.

6. AKIDA 2.0 TENNS IP - not a product still in development.

7. TENNS software - not a product still in development.

8. TENNS Pleiades software - not a product still in development.

9. VVDN AKIDA Edge Box – It's not intended to be a source of significant revenue but rather to showcase Akida's capabilities.

10. EDGX-1 Brain – This is not a product; it is a partnership project being undertaken under a non-binding Memorandum of Understanding with EDGX.

11. (To be released) Unigen AKIDA Ai Cupcake Edge Server – as per VVDN box

12. (Under Development) Cloud based AKIDA FPGA Development Environment.

13. Models for Noise Cancellation and Keyword Spotting.

14. Optimised models for GENAi applications at the Edge including ASR.

Items 12 to 14 are not products but are under development to support the IP being sold. Sean has repeatedly emphasised that we are an IP-focused company. Our current IP product portfolio available now is AKIDA1.0, AKIDA1500, and AKIDA 2.0, and these are the products we are aiming to sell to reach viability.

Once again, a most impressive and unbiased factual summation of the current state of affairs concerning our various patented technologies.

 

[Xray1]

Your not helpful at all

Obviously, you are either unable or unwilling to accept the realities and factual information of what AI-Inquirer has provided and detailed in his post. I note your response post back to AI-Inquirer stating.. "Your not helpful at all" ... then, why don't you take the time out to detail each of his points made as to why you think they are unhelpful...

It's about time, that some posters here should take off their rose coloured glasses and refrain from being somewhat Co stooges.

 

[DingoBorat]

Once again, a most impressive and unbiased factual summation of the current state of affairs concerning our various patented technologies.

Except, I think @AI_Inquirer, is mistaken about the following not being "products".

"6. AKIDA 2.0 TENNS IP - not a product still in development"

"7 TENNS software - not a product still in development"

"8. TENNS Pleiades software - not a product still in development"


While "still being developed" as are all BrainChip technologies (nothing remains "fixed" when it come to high technology).

The language from BrainChip, tells me, that they are ready to be utilised, as they stand now.

Introducing TENN: Revolutionizing Computing with an Energy Efficient Transformer Replacement

Introducing TENN: Revolutionizing Computing with an Energy Efficient Transformer Replacement By Dr. Tony Lewis, CTO at BrainChip, Olivier Coenen, Senior Research Scientist at BrainChip and Yan Ru Pei, research Scientist at BrainChip The Rise of TENN: A Game Changer in AI TENN, or...

brainchip.com

"The implementation of TENN within BrainChip’s hardware in the Akida 2.0, showcases a significant step forward in hardware-accelerated AI. Akida 2.0’s architecture is designed to fully exploit TENN’s capabilities, featuring a mesh network of nodes each equipped with an event-based TENN processing unit. This design ensures scalability and enhances computational efficiency, making it suitable for deployment in environments where power and space are limited"

BrainChip Introduces TENNs-PLEIADES in New White Paper

BrainChip Introduces TENNs-PLEIADES in New White Paper Laguna Hills, Calif. – June 5, 2024 – BrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), the world’s first commercial producer of ultra-low power, fully digital, event-based, neuromorphic AI, today released a white paper...

brainchip.com

"TENNs-PLEIADES is the latest technological advancement added to BrainChip’s IP portfolio and an expansion of Temporal Event-Based Neural Nets (TENNs), the company’s approach to streaming and sequential data. It complements the company’s neural processor, Akida™ IP, an event-based technology that is inherently lower power when compared to conventional neural network accelerators. Lower power affords greater scalability and lower operational costs"


Also, I don't think there is an "AKIDA 1.5 IP" and that AKD1500, comes under AKIDA 1.0 IP.

I'm happy to be corrected.

Maybe I "am" wearing these?...

 

[manny100]

Once again, a most impressive and unbiased factual summation of the current state of affairs concerning our various patented technologies.

Its actually a very positive post. It highlights the strength of our portfolio which underpins the value of BRN in the absence of deals which will materialize in due course.

 

[Diogenese]

hi @Diogenese .................. like to ask you if you would agree that SNR (signal to noise ratio) could be seen as the same principals as SNNs ?
And would the use of SNNs give a higher SNR?
Appreciate your thoughts.
Cheers

Hi mrgds,

There is a similarity in that we are talking about selecting correct signals from erroneous signals, but analog NNs, which rely on the amplitude of signals to convey information, suffer from manufacturing variability which means that a signal applied to two different circuits could produce different amplitudes outputs. Given that there are thousands of operations involved, the error can be significant.

SNR is more a problem of selecting a signal from a lot of background "static". This would be more apposite for noise-cancelling systems.

Analog SNNs are about variations in the inherent value (amplitude) of the signal rather than selecting signals from extraneous noise.

Digital uses a simple ON/OFF switch, which has more than enough tolerance to override manufacturing variability.

As a rudimentary example, where 10 analog signals are combined in a 5v system with 0.2v manufacturing variability per device, the cumulative error could be 2v (= 40%), whereas, in a 5v digital system, the switching threshold could be set at 3v, which would easily accommodate the 0.2v variation.

 

[Diogenese]

1. AKD1000 SoC – A limited run of chips was produced with no plans for further production. The chips did not generate significant revenue or sell and are now being repurposed for applications like edge boxes.

2. AKD1000 IP – Two IP licenses have been sold (Renesas and Megaship), and Sean mentions that ongoing engagement continues with others.

3. AKD1500 Chip - The AKD1500 is an accelerator reference chip, which assists partners in developing and demonstrating their solutions as a stepping stone to integrating the Akida IP into their production SoCs - It’s not meant to be a revenue stream (just like edge boxes).

4. AKD1500 IP – No licenses have been sold. However, it's worth noting that Megachip said on LinkedIn they played a role in developing Akida1500, which I found interesting, as it suggested they might have had an end client or their own use in mind.

5. AKIDA 2.0 IP – No licenses sold.

6. AKIDA 2.0 TENNS IP - not a product still in development.

7. TENNS software - not a product still in development.

8. TENNS Pleiades software - not a product still in development.

9. VVDN AKIDA Edge Box – It's not intended to be a source of significant revenue but rather to showcase Akida's capabilities.

10. EDGX-1 Brain – This is not a product; it is a partnership project being undertaken under a non-binding Memorandum of Understanding with EDGX.

11. (To be released) Unigen AKIDA Ai Cupcake Edge Server – as per VVDN box

12. (Under Development) Cloud based AKIDA FPGA Development Environment.

13. Models for Noise Cancellation and Keyword Spotting.

14. Optimised models for GENAi applications at the Edge including ASR.

Items 12 to 14 are not products but are under development to support the IP being sold. Sean has repeatedly emphasised that we are an IP-focused company. Our current IP product portfolio available now is AKIDA1.0, AKIDA1500, and AKIDA 2.0, and these are the products we are aiming to sell to reach viability.

cc @DingoBorat

Hi AI_I,

Mercedes and Valeo are at least 2 EAPs we have been strongly involved with for a few years. We know that, in their upcoming products, both are planning to use software for signal processing.

Both would have had access to Akida 2/TeNNs simulation software since the filing of the TeNNs patent over 2 years ago.

Since TeNNs is still in development, they would be reluctant to commit to silicon, so the absence of Akida 2/TeNNs silicon from SCALA 3 and Mercedes DMS for example is not surprising. Software can be continually updated.

Given the previously expressed enthusiasm for Akida by both companies, I am hopeful that they are using the Akida 2/TeNNs software for signal processing. Only yesterday I read a post here (@Tuliptrader ?) about a linkedin a post by a senior MB exec in charge of MBOS, which referred to end-to-end-NNs.

As I've said a few times now, I suspect that both Valeo and Mercedes are using Akida2/TeNNs software for signal processing in their new releases until the development of TeNNs has plateaued and been proven at a satisfactory performance level. If this is the case, I would expect that there will be a commercial licence, albeit under NDA.

Also, software NNs are not as big a problem with ICEs as they are with EVs,

However, such speculation is notoriously unreliable and should not form the basis if investment decisions.

 

[mrgds]

Hi mrgds,

There is a similarity in that we are talking about selecting correct signals from erroneous signals, but analog NNs, which rely on the amplitude of signals to convey information, suffer from manufacturing variability which means that a signal applied to two different circuits could produce different amplitudes outputs. Given that there are thousands of operations involved, the error can be significant.

SNR is more a problem of selecting a signal from a lot of background "static". This would be more apposite for noise-cancelling systems.

Analog SNNs are about variations in the inherent value (amplitude) of the signal rather than selecting signals from extraneous noise.

Digital uses a simple ON/OFF switch, which has more than enough tolerance to override manufacturing variability.

As a rudimentary example, where 10 analog signals are combined in a 5v system with 0.2v manufacturing variability per device, the cumulative error could be 2v (= 40%), whereas, in a 5v digital system, the switching threshold could be set at 3v, which would easily accommodate the 0.2v variation.

Hi "Dodgyknees"
Thanks very much for the reply, and the time you put into your response.
Helps making it a little "less muddy" for noobs like myself.
Cheers

Akida Ballista

 

[Rach2512]

View attachment 68081

BrainChip Podcast Epi 34: The State of Neuromorphic Computing

In this episode of the “This is Our Mission” podcast, Sean Hehir interviews Dr. Eric Gallo, a Senior Principal at Accenture Labs. They discuss the advantages of neuromorphic technology and its impact on edge computing, as well as advancements in SpaceTech.

www.wevolver.com

BrainChip Podcast Epi 34: The State of Neuromorphic Computing​

In this episode of the “This is Our Mission” podcast, Sean Hehir interviews Dr. Eric Gallo, a Senior Principal at Accenture Labs. They discuss the advantages of neuromorphic technology and its impact on edge computing, as well as advancements in SpaceTech.​

Hiba Akbar
15 Aug, 2024. 4 min read
Follow

Topic
A.I.
Follow
Tags
artificial intelligence
Edge AI
neuromorphic computing
space
Accenture is at the forefront of technological innovation with a focus on developing next-generation computing technologies. In a recent podcast, Dr. Eric Gallo, a senior principal at Accenture Labs, shared insights into the promising field of neuromorphic computing. This technology mimics how the human brain processes information and offers significant advantages in power efficiency and real-time data processing.
As industries increasingly rely on edge computing, neuromorphic systems present a unique solution to the challenges of energy consumption and integration in smart devices. This episode of the BrainChip podcast explores Accenture's initiatives in neuromorphic computing. They also explore its applications in various sectors and the potential it holds for the future of edge intelligence.

The Rise of Edge Computing and the Need for Heterogeneous Computing​

Edge computing is rapidly gaining traction as industries increasingly rely on real-time data processing and intelligence at the edge of the network. As the industry matures, there is a growing understanding that a heterogeneous set of computing devices is necessary to achieve optimal results.
Traditional computing architectures will continue to play a role, but they will make way for other specialized architectures that excel in different situations and locations. This will create a continuum where edge architectures, cloud architectures, and other specialized architectures work together to provide the right amount of compute power where it's needed most.
Companies with edge devices are actively examining their AI strategies to avoid being left behind. While computing power is readily available, the ability to perform inference very close to the edge is a key focus for many organizations. This is where neuromorphic computing shines, offering a low-power solution for real-time data processing at the edge. This energy efficiency is not just a minor improvement; it’s a game-changer.
Dr. Gallo highlighted how neuromorphic technology could potentially achieve power savings of up to 100,000 times compared to current methods. This opens up new possibilities for developing intelligent devices that can operate in environments where power is limited, such as in remote locations, wearable devices, or even deep space missions. Neuromorphic computing is still in its early stages, but its potential to transform industries is immense.

The Power of Neuromorphic Technology​

One of the most significant advantages of neuromorphic technology is its remarkable energy efficiency. Traditional computing systems often require a large amount of power to perform complex tasks. However, neuromorphic systems can achieve the same results while using only a fraction of that energy. This makes them ideal for applications where power is a critical concern.
Dr. Eric Gallo explained that this technology could be a game-changer in various fields. For example, in defense, neuromorphic chips could be used to create advanced situational awareness systems for soldiers. These systems could process vast amounts of data in real time, helping soldiers make better decisions in the field without the need for bulky, power-hungry equipment.
Neuromorphic technology could also enhance the intelligence of factory equipment in industrial settings. Machines equipped with neuromorphic chips could adapt to changing conditions on the fly, improving efficiency and reducing downtime. These chips can also be powered by small batteries or energy harvesters, making them suitable for environments where access to power is limited.
As this technology continues to develop, its impact on different sectors will also grow. This will lead to more efficient and intelligent solutions.

Accenture's Neuromorphic Computing Initiatives in Space​

Eric Gallo believes that neuromorphic architectures can enable smart satellites and other space devices without the significant power and thermal constraints of traditional computing systems.
Accenture is working towards demonstrating real-time neuromorphic computing in space. By leveraging the low-power capabilities of neuromorphic chips, Accenture aims to make space devices more intelligent and responsive.
The space industry has traditionally relied on less advanced computing technologies due to the challenges of power and heat dissipation. However, with the emergence of neuromorphic computing, there is a sudden realization that space systems can be made much smarter without the usual constraints.

Accenture's Partnership with BrainChip​

Accenture has formed a strong partnership with BrainChip, a leader in neuromorphic computing technology. This collaboration uses the neuromorphic chip to explore practical applications in various industries. Dr. Eric Gallo, who leads Accenture's neuromorphic initiatives, has shared valuable insights from their work with this advanced technology.
The chip stands out for its ability to save power compared to traditional computing systems. In practical tests, Accenture observed that systems using the chip consumed only a fraction of the power—about one-fifth—compared to conventional setups. This efficiency is important for applications where power resources are limited, such as edge devices and satellites.
Working with BrainChip has allowed Accenture to access cutting-edge neuromorphic technology and gain practical experience in real-world environments. The partnership has been characterized by strong support and collaboration that enables both teams to tackle challenges and continuously improve their systems.

The Future of Neuromorphic Computing: Spanning Material Spaces and Scales​

One of the most exciting aspects of neuromorphic computing is its potential to span a wide range of material spaces and scales. Dr. Gallo envisions the possibility of creating tiny, biodegradable neuromorphic sensors that can be used in applications like water quality monitoring. These sensors could transmit data to small neural networks, which could then determine if there is a need for concern.
At the other end of the spectrum, neuromorphic architectures are enabling large-scale neural networks like Spike GPT. These advanced systems demonstrate the versatility of neuromorphic computing, which can be applied from the smallest sensors to the most powerful artificial intelligence systems.
Dr. Gallo emphasizes that anyone, even those without expertise in computing architectures, can get involved and contribute to the advancement of neuromorphic computing. The field is open to new ideas and innovations, and there are many opportunities for individuals to make meaningful contributions.

Final Words​

Looking ahead, the future of neuromorphic computing appears bright. Its ability to span a wide range of materials and scales opens doors for anyone interested in contributing to this exciting field. As organizations like Accenture lead the charge, we can expect to see more practical implementations that harness the power of neuromorphic technology, ultimately making our world smarter and more efficient. To learn more about the future of neuromorphic computing watch the full podcast above!

Are we in a partnership with Accenture! I knew we had done podcasts but had it been mentioned that we were in a partnership, I must have fallen asleep if I missed this?