BRN Discussion Ongoing

[Fact Finder]

Hey all, apologies if already been posted but I couldn't see it in the 3 seconds of searching I performed
Looks like the results were published in Nov 22.
Obviously not Akida but a great read on what NM tech is capable of.

Neuromorphic semiconductor device achieves world's highest handwriting pattern recognition rate

Neuromorphic semiconductor device achieves world's highest handwriting pattern recognition rate​

by National Research Council of Science & Technology

A diagram of battery material-incorporated next-generation neuromorphic semiconductor device with high-density and high-reliability (left) Photo and handwriting pattern recognition accuracy of a 3-terminal-based device with high density made of a thin film of lithium-ion (right). Credit: Korea Institute of Materials Science (KIMS)
A research team led by Dr. Yong-hun Kim and Dr. Jeong-Dae Kwon has successfully developed the world's first neuromorphic semiconductor device with high-density and high-reliability by developing a thin film of lithium-ion battery materials. They achieved this by producing ultra-thin lithium ions, a key material of lithium-ion batteries that have been in the spotlight recently, and combining it with two-dimensional nano-materials. The research team is from the Surface & Nano Materials Division at the Korea Institute of Materials Science (KIMS).

A neuromorphic semiconductor device has synapses and neurons similar to the human brain, which processes and memorizes information. The synaptic device receives signals from neurons and modulates the synaptic weight (connection strength) in various ways to simultaneously process and store information. In particular, the linearity and symmetry of synaptic weights enables various pattern recognition with low power.
Traditional methods for controlling synaptic weights use charge traps between interfaces of heterogeneous materials or oxygen ions. In this case, however, it is difficult to control the movement of ions in the desired direction according to the external electric field. The researchers solved this problem with an artificial intelligence semiconductor device with high density by developing a thin film process while maintaining the mobility of lithium ions according to the external electric field. The thin film—with a thickness of several tens of nanometers—enables fine pattern processing while controlling the thickness of the wafer scale.
The research team developed the film using a vacuum sputtering deposition method that is used in general semiconductor processing. The thickness of the deposited film is less than 100 nanometers. After a transistor-type device on a silicon wafer substrate is manufactured and when an electric field is applied from the outside, the lithium ions in the charged lithium thin film move so that the conductivity of the channel can be precisely controlled.
The research team implemented an artificial neural network learning pattern using this synapse device and developed handwriting image pattern recognition. The artificial intelligence semiconductor device shows a handwriting pattern recognition rate of about 96.77% by maintaining the characteristics of finely adjusted synaptic weight even in an electric field repeated more than 500 times.
The research team stated, "Our next-generation neuromorphic semiconductor device does not require CPU and memory, the traditional Von Neumann-type information processing device and information storage device. It can simultaneously process and store information and learns and recognizes images such as handwriting patterns. It is expected to be applied to various low-power artificial intelligence devices such as world-class neuromorphic hardware systems, haptic devices, and vision sensors."
The results of the research were published in ACS Applied Materials & Interfaces on November 17, 2022. The research team is conducting follow-up research on low-power artificial intelligence devices and wearable edging devices

Well beyond the pay grade of this old technophobe but just wondering what its tolerance to heat and cold is and if radiation will punch holes through this very thin material and create a new swiss cheese semiconductor in extreme use cases. If so they can kiss goodbye the automotive, mining, heavy industry, defence and space applications. Then they have the problem of foundries deciding to take up and run with this technology when current technology like 22nm at Global Foundries is making science fiction redundant. Might be quite a long time before it becomes a necessity.

My opinion only DYOR
FF

AKIDA BALLISTA

 

[misslou]

Oh the vibe is so much better in here now!!!!

 

[BaconLover]

 

[Rskiff]

Oh the vibe is so much better in here now!!!!

And so it should be with yesterdays podcast! Peter set out a truely amazing road map of where he intends to drive development of Akida. Also having FF back sparks more positivity here as well.

 

[Space Cadet]

According to ChatGPT:

"The shadowy figure is most likely to be the rival real estate agent trying to break into the area and attempting to destroy Browns reputation."

And when they check the crap for DNA and identify the culprit, they can also hook it up to Akida and let him know he has bowel cancer.

SC

 

[TECH]

It seems very clear to me at least that the Boards business plan is being rolled out beautifully.

I have had a number of shareholders comment to me privately that they aren't too impressed with Sean's apparent silence as our CEO,
hearing from him once each year at an AGM doesn't cut it with them, yes he comments in the quarterlies and in Annual Reports, I reserve
my opinion at this point, I guess we don't wish to hear fluff throughout the year either, only solid facts and figures.

My insider information (Bell Direct) tells me that the Annual Report will be filed tomorrow, time shall tell, no matter what it contains, we,
Brainchip are definitely moving forward, and don't we love it!

"Right Place.....Right Time"

Yesterday I was glad Peter said Albert Einstein and not Young Einstein

Texta

 

[VictorG]

According to RT, BRN will be located at booth 2-238 at the Embedded World 2023.
According to Embedded World's website, a company called AIZIP is in booth 2-238

AIZIP doesn't list BRN as partners but they do say they have new partnerships to be announced shortly.
Interestingly though AIZIP's link to their partnership with Renesas describes everything Akida does.

Partner Solution - Renesas - Person Access System - Aizip

aizip.ai

Would anyone with ChatGPT ask it the following question.
Who are the shareholders of AIZIP Inc.

 

[Diogenese]

Hey all, apologies if already been posted but I couldn't see it in the 3 seconds of searching I performed
Looks like the results were published in Nov 22.
Obviously not Akida but a great read on what NM tech is capable of.

Neuromorphic semiconductor device achieves world's highest handwriting pattern recognition rate

Neuromorphic semiconductor device achieves world's highest handwriting pattern recognition rate​

by National Research Council of Science & Technology

A diagram of battery material-incorporated next-generation neuromorphic semiconductor device with high-density and high-reliability (left) Photo and handwriting pattern recognition accuracy of a 3-terminal-based device with high density made of a thin film of lithium-ion (right). Credit: Korea Institute of Materials Science (KIMS)
A research team led by Dr. Yong-hun Kim and Dr. Jeong-Dae Kwon has successfully developed the world's first neuromorphic semiconductor device with high-density and high-reliability by developing a thin film of lithium-ion battery materials. They achieved this by producing ultra-thin lithium ions, a key material of lithium-ion batteries that have been in the spotlight recently, and combining it with two-dimensional nano-materials. The research team is from the Surface & Nano Materials Division at the Korea Institute of Materials Science (KIMS).

A neuromorphic semiconductor device has synapses and neurons similar to the human brain, which processes and memorizes information. The synaptic device receives signals from neurons and modulates the synaptic weight (connection strength) in various ways to simultaneously process and store information. In particular, the linearity and symmetry of synaptic weights enables various pattern recognition with low power.
Traditional methods for controlling synaptic weights use charge traps between interfaces of heterogeneous materials or oxygen ions. In this case, however, it is difficult to control the movement of ions in the desired direction according to the external electric field. The researchers solved this problem with an artificial intelligence semiconductor device with high density by developing a thin film process while maintaining the mobility of lithium ions according to the external electric field. The thin film—with a thickness of several tens of nanometers—enables fine pattern processing while controlling the thickness of the wafer scale.
The research team developed the film using a vacuum sputtering deposition method that is used in general semiconductor processing. The thickness of the deposited film is less than 100 nanometers. After a transistor-type device on a silicon wafer substrate is manufactured and when an electric field is applied from the outside, the lithium ions in the charged lithium thin film move so that the conductivity of the channel can be precisely controlled.
The research team implemented an artificial neural network learning pattern using this synapse device and developed handwriting image pattern recognition. The artificial intelligence semiconductor device shows a handwriting pattern recognition rate of about 96.77% by maintaining the characteristics of finely adjusted synaptic weight even in an electric field repeated more than 500 times.
The research team stated, "Our next-generation neuromorphic semiconductor device does not require CPU and memory, the traditional Von Neumann-type information processing device and information storage device. It can simultaneously process and store information and learns and recognizes images such as handwriting patterns. It is expected to be applied to various low-power artificial intelligence devices such as world-class neuromorphic hardware systems, haptic devices, and vision sensors."
The results of the research were published in ACS Applied Materials & Interfaces on November 17, 2022. The research team is conducting follow-up research on low-power artificial intelligence devices and wearable edging devices

Sort of MemRistor - don't know how it would go for radhardness.

 

[HarryCool1]

It seems very clear to me at least that the Boards business plan is being rolled out beautifully.

I have had a number of shareholders comment to me privately that they aren't too impressed with Sean's apparent silence as our CEO,
hearing from him once each year at an AGM doesn't cut it with them, yes he comments in the quarterlies and in Annual Reports, I reserve
my opinion at this point, I guess we don't wish to hear fluff throughout the year either, only solid facts and figures.

My insider information (Bell Direct) tells me that the Annual Report will be filed tomorrow, time shall tell, no matter what it contains, we,
Brainchip are definitely moving forward, and don't we love it!

"Right Place.....Right Time"

Yesterday I was glad Peter said Albert Einstein and not Young Einstein

Texta

Whoa, Kemosabe! If the movie is to be believed, he contributed immensely!!

 

[Cirat]

Firstly if others want to go back and check I warned against the euphoria that some greeted the admission of Brainchip into the ASX200. (I warned all at the time it would bring to the company a new level of short activity. I suggested this would not be good for the share price and short term investors.)

And it did.

Early in the HOT Ukraine war I warned that this war was going to crush economic activity in the EU, UK and as a result hurt the US at a time when it was flexing its muscles in a trade war with China.

And it did.

I also stated that all these Global storms would however work to Australia’s advantage. I said China needed Australia more than Australia needed China. I said that Europe & the UK needed Australia to fill in the energy gap left by Russia destroying for decades to come any thoughts that Europeans and British peoples had that Russia could be a reliable trading partner. I also pointed to the unsavoury fact that Brainchip would benefit from the resultant increase in military spending by the US and its allies.

And it did.

I said many times the lucky country would continue to prove it was the Economic Miracle.

And it did.

While Brainchip is an Australian born company like Atlassian it too is not a company that finds its success in Australia. As the CEO Sean Hehir said at the AGM shareholders need ‘to get their heads out of their backsides’ and accept that Brainchip is an international company. (He did not use ‘their heads out of their backsides’ line but this was his intent.) Brainchip’s future is international and as such it is subject to all the economic and trading headwinds that circulate in the international economy. It cannot shelter from these forces by hiding away in the economic miracle and become a technology giant. It has to go international.

And it did.

Against this background of increased short activity by ASX200 practitioners including activist shorts, global supply chain disruption, trade wars involving Russia, China, USA and Western countries Brainchip warned of a slow down in the markets it inhabits. Despite these factors Brainchip pushes to grow its ecosystem and increase its sales and marketing efforts across the globe.

And it did.

As for sales cycles I pointed firstly to Prophesee and Sony and it’s shortly to be released mass market vision sensor. Why because it is a perfect example of a neuromorphic product being sold to and developed by a non neuromorphic global company. The evidence is clear Prophesee and Sony commenced working together in 2020. It is now 2023. Time to market at least three years. Despite this great engagement Prophesee raised $50 million in capital late last year to keep the doors open. Substantial income clearly unlikely to start to flow from Sony to Prophesee before end 2023.

But it will.

Renesas announced by Brainchip as an IP purchaser 23.12.20. Renesas announces tape out of AKIDA powered chip just recently with mass production late 2023. Income will flow in 2024.

And it will.

Socionext VP of Sales committed to assisting Brainchip to commercialise the AKIDA product family and just a month or so before the full commercial release of the AKD1000 in 2021 Socionext commenced to promote AKD1000 at trade fairs using the AKD1000 engineering samples as demonstrators. In January, 2023 Socionext announced at CES that they have a range of AKIDA based automotive solutions for customer consideration.

And they are.

Mercedes Benz revealed they were working with Brainchip at CES2022 with vehicles coming to market in 2024 and Brainchip have stated they continue to work with Mercedes Benz to achieve this end.

And they will.

Nviso struggled to find traction in the market place for 11 years and now having discovered Brainchip’s AKIDA solution is finally able to realise its ambition.

And it will.

Prophesee was building a glass house until it was completed by Brainchip’s AKIDA after having first contact with Adam Osseiran 6 years ago is looking towards a commercial engagement.

And it will.

Intel, ARM & SiFive having been announced over 2022 are engaged in marketing Brainchip’s AKIDA solution along with Edge Impulse.

And they are.

MegaChips having first discovered Brainchip’s AKIDA technology solution in 2020 set about creating a section within its organisation of engineers trained in and dedicated to the use and implementation of the AKIDA technology solution and in late 2021 they with Brainchip announced the purchase of a full AKIDA IP licence and set about marketing AKIDA based solutions to their customer base.

And they do.

So just like the Little Red Hen who found some grain and wanted to make bread despite the criticism of the other animals in the barnyard who are lazy or ignorant and not prepared to suffer the hard work and delays involved in taking wheat through the process necessary to become bread Brainchip just keeps forging ahead in an industry where from first customer contact to product on shelf is three years plus and where paradigm shifting is slightly more difficult than alchemy.

My opinion only DYOR
FF

AKIDA BALLISTA

Apologies to the classically trained but I am reading to an audience of under 2 year olds where The Little Red Hen rules.

Welcome back FF !!

I think it was Tech (sorry if it was someone else) that recommended the book "Culture Won" about ARM.

I purchased the book and find it an incredible read. The similarities are amazing and give you a greater understanding of; why various paths to market are taken, time to financial reward, importance of ecosystem partnerships, partners input and many more... especially culture.

I definitely recommend this book to better understand our own unfolding story.

Patience can be tested for so many individual reasons but I believe we are stronger than ever.

GLTA

 

[Fullmoonfever]

Looks like this Team want to explore us a little further

Haven't looked into who or what ANR Project DeepSee is yet - that's where the grant is coming from

Evaluation of neuromorphic AI with embedded Spiking Neural Networks – LEAT

• DECEMBER 10, 2022
• COMMENTS OFF

Evaluation of neuromorphic AI with embedded Spiking Neural Networks​

Background
AI is proliferating everywhere even to embedded systems to integrate intelligence closer to the sensors (IoT, drones, vehicles, satellites ...). But the energy consumption of current Deep learning solutions makes classical AI hardly compatible with energy and resource constrained devices. Edge AI is a recent subject of research that needs to take into account the cost of the neural models both during the training and during the prediction. An original and promising solution to face these constraints is to merge compression technics of deep neural networks and event-based encoding of information thanks to Spiking neural networks (SNN). SNN are considered as third generation of artificial neural networks and are inspired from the way the information is encoded in the brain, and previous works tend to
conclude that SNN are more efficient than classical deep networks [3]. This internship project aims at confirming this assumption by converting classical CNN to SNN from standard Machine Learning frameworks (Keras) and deploy the resulting neural models onto the Akida neuromorphic processor from BrainChip company [4]. The results obtained in terms of accuracy, latency and energy will be compared to other existing embedded solutions for Edge AI [2].
Project mission
The project mission will be organized in several periods:

• Bibliographic study on spiking neural network training
• Introduction to the existing Sw framework from BrainChip
• Training of convolutional neural networks for embedded applications [1] and conversion from CNN to SNN from Keras
• Deployment of the SNN onto Akida processing platform
• Experiments and measurements
• Publication in an international conference.

References
[1] L Cordone, Miramond B, Thierion, Object Detection with Spiking Neural Networks on Automotive Event Data, IEEE International Joint Conference on Neural Networks (IJCNN), 2022[2] N Abderrahmane, Miramond B, Kervennic E, A Girard, SPLEAT: SPiking Low-power Event-based
ArchiTecture
for in-orbit processing of satellite imagery, IEEE International Joint Conference on Neural Networks, 1-10, 2022

[3] E Lemaire, L Cordone, A Castagnetti, PE Novac, J Courtois, B Miramond, An Analytical Estimation of Spiking
Neural Networks Energy Efficiency, Springer International Conference on Neural Information Processing, 2022[4] T. Álvarez-Sánchez, et al, Detection of facial emotions using neuromorphic computation, Applications of
Digital Image Processing, 2022

Practical information
Location : LEAT Lab / SophiaTech Campus, Sophia Antipolis
Duration : 6 months from march 2023
Grant : from ANR project DeepSee
Profile : Machine learning, Artificial intelligence, Artificial neural networks, Python, Keras, Pytorch
Research keywords : Spiking neural network, Edge AI, neuromorphic computing

Contact and supervision
Benoît Miramond, Andrea Castagnetti
LEAT Lab – University Cote d'Azur / CNRS
Polytech Nice Sophia
04.89.15.44.39. / benoit.miramond@univ-cotedazur.fr

Le LEAT – LEAT

leat.univ-cotedazur.fr

THE LEAT​

• Home/
• The LEAT

Presentation​

The Laboratory of Electronics, Antennas and Telecommunications (LEAT) is a Joint Unit Université Côte d'Azur – CNRS (UMR n°7248). It is located on the SophiaTech campus, which is a training and research center dedicated to Information and Communication Technologies (ICT) involving academic actors (UNS, INRIA, EURECOM, CNRS, Polytech'Nice Sophia, Mines Paris Tech, etc.), competitiveness clusters, numerous associations and technological platforms.

Management Team​

Director: Robert Staraj (PR, UCA)
Deputy Directors: Jean-Marc Ribero (PR, UCA) and François Verdier (PR, UCA)
Administrative Manager: Françoise Trucas (CNRS)

Research activities​

Research activities are carried out in the field of telecommunications, radar, e-health, security, smart buildings, earth observation, sustainable development, etc. They are organized into three themes: EDGE (Edge Computing and Digital Systems), CMA (Antenna Design and Modeling) and ISA (Microwave Imaging and Antenna Systems).

Environment​

LEAT participates in the activities of the main competitiveness clusters in the region. It is associated with the research program of the "Laboratory of Excellence" Labex UCN@Sophia. Among the partners with which the laboratory is associated, are present the two supervisory bodies (Nice Sophia Antipolis University and the CNRS), but also Eurecom, Inria and the I3S and LTCI laboratories. The Labex is part of the IDEX UCAJEDI.
The Centre de RErecherche Mutualisé sur les ANTennes (CREMANT), created in 2008, is a joint laboratory between the University of Nice Sophia Antipolis (now Université Côte d'Azur), the CNRS and Orange. It allowed the pooling of personnel and equipment between LEAT academic researchers and Orange Labs La Turbie engineers on common research topics (antenna integration, engineering for e-health, multisensor, MIMO and massive MIMO systems, antennas based on new materials, electromagnetic modeling).

 

[Fullmoonfever]

Looks like this Team want to explore us a little further

Haven't looked into who or what ANR Project DeepSee is yet - that's where the grant is coming from

Evaluation of neuromorphic AI with embedded Spiking Neural Networks – LEAT

• DECEMBER 10, 2022
• COMMENTS OFF

Evaluation of neuromorphic AI with embedded Spiking Neural Networks​

Background
AI is proliferating everywhere even to embedded systems to integrate intelligence closer to the sensors (IoT, drones, vehicles, satellites ...). But the energy consumption of current Deep learning solutions makes classical AI hardly compatible with energy and resource constrained devices. Edge AI is a recent subject of research that needs to take into account the cost of the neural models both during the training and during the prediction. An original and promising solution to face these constraints is to merge compression technics of deep neural networks and event-based encoding of information thanks to Spiking neural networks (SNN). SNN are considered as third generation of artificial neural networks and are inspired from the way the information is encoded in the brain, and previous works tend to
conclude that SNN are more efficient than classical deep networks [3]. This internship project aims at confirming this assumption by converting classical CNN to SNN from standard Machine Learning frameworks (Keras) and deploy the resulting neural models onto the Akida neuromorphic processor from BrainChip company [4]. The results obtained in terms of accuracy, latency and energy will be compared to other existing embedded solutions for Edge AI [2].
Project mission
The project mission will be organized in several periods:

• Bibliographic study on spiking neural network training
• Introduction to the existing Sw framework from BrainChip
• Training of convolutional neural networks for embedded applications [1] and conversion from CNN to SNN from Keras
• Deployment of the SNN onto Akida processing platform
• Experiments and measurements
• Publication in an international conference.

References
[1] L Cordone, Miramond B, Thierion, Object Detection with Spiking Neural Networks on Automotive Event Data, IEEE International Joint Conference on Neural Networks (IJCNN), 2022[2] N Abderrahmane, Miramond B, Kervennic E, A Girard, SPLEAT: SPiking Low-power Event-based
ArchiTecture
for in-orbit processing of satellite imagery, IEEE International Joint Conference on Neural Networks, 1-10, 2022

[3] E Lemaire, L Cordone, A Castagnetti, PE Novac, J Courtois, B Miramond, An Analytical Estimation of Spiking
Neural Networks Energy Efficiency, Springer International Conference on Neural Information Processing, 2022[4] T. Álvarez-Sánchez, et al, Detection of facial emotions using neuromorphic computation, Applications of
Digital Image Processing, 2022

Practical information
Location : LEAT Lab / SophiaTech Campus, Sophia Antipolis
Duration : 6 months from march 2023
Grant : from ANR project DeepSee
Profile : Machine learning, Artificial intelligence, Artificial neural networks, Python, Keras, Pytorch
Research keywords : Spiking neural network, Edge AI, neuromorphic computing

Contact and supervision
Benoît Miramond, Andrea Castagnetti
LEAT Lab – University Cote d'Azur / CNRS
Polytech Nice Sophia
04.89.15.44.39. / benoit.miramond@univ-cotedazur.fr

Le LEAT – LEAT

leat.univ-cotedazur.fr

THE LEAT​

• Home/
• The LEAT

Presentation​

The Laboratory of Electronics, Antennas and Telecommunications (LEAT) is a Joint Unit Université Côte d'Azur – CNRS (UMR n°7248). It is located on the SophiaTech campus, which is a training and research center dedicated to Information and Communication Technologies (ICT) involving academic actors (UNS, INRIA, EURECOM, CNRS, Polytech'Nice Sophia, Mines Paris Tech, etc.), competitiveness clusters, numerous associations and technological platforms.

Management Team​

Director: Robert Staraj (PR, UCA)
Deputy Directors: Jean-Marc Ribero (PR, UCA) and François Verdier (PR, UCA)
Administrative Manager: Françoise Trucas (CNRS)

Research activities​

Research activities are carried out in the field of telecommunications, radar, e-health, security, smart buildings, earth observation, sustainable development, etc. They are organized into three themes: EDGE (Edge Computing and Digital Systems), CMA (Antenna Design and Modeling) and ISA (Microwave Imaging and Antenna Systems).

Environment​

LEAT participates in the activities of the main competitiveness clusters in the region. It is associated with the research program of the "Laboratory of Excellence" Labex UCN@Sophia. Among the partners with which the laboratory is associated, are present the two supervisory bodies (Nice Sophia Antipolis University and the CNRS), but also Eurecom, Inria and the I3S and LTCI laboratories. The Labex is part of the IDEX UCAJEDI.
The Centre de RErecherche Mutualisé sur les ANTennes (CREMANT), created in 2008, is a joint laboratory between the University of Nice Sophia Antipolis (now Université Côte d'Azur), the CNRS and Orange. It allowed the pooling of personnel and equipment between LEAT academic researchers and Orange Labs La Turbie engineers on common research topics (antenna integration, engineering for e-health, multisensor, MIMO and massive MIMO systems, antennas based on new materials, electromagnetic modeling).

Part of the same research by the looks. Oct 22.

https://arxiv.org/pdf/2210.13107.pdf

 

[equanimous]

Mic drop.

 

[Fullmoonfever]

Part of the same research by the looks. Oct 22.

https://arxiv.org/pdf/2210.13107.pdf

Looks like this Team want to explore us a little further

Haven't looked into who or what ANR Project DeepSee is yet - that's where the grant is coming from

Evaluation of neuromorphic AI with embedded Spiking Neural Networks – LEAT

• DECEMBER 10, 2022
• COMMENTS OFF

Evaluation of neuromorphic AI with embedded Spiking Neural Networks​

Background
AI is proliferating everywhere even to embedded systems to integrate intelligence closer to the sensors (IoT, drones, vehicles, satellites ...). But the energy consumption of current Deep learning solutions makes classical AI hardly compatible with energy and resource constrained devices. Edge AI is a recent subject of research that needs to take into account the cost of the neural models both during the training and during the prediction. An original and promising solution to face these constraints is to merge compression technics of deep neural networks and event-based encoding of information thanks to Spiking neural networks (SNN). SNN are considered as third generation of artificial neural networks and are inspired from the way the information is encoded in the brain, and previous works tend to
conclude that SNN are more efficient than classical deep networks [3]. This internship project aims at confirming this assumption by converting classical CNN to SNN from standard Machine Learning frameworks (Keras) and deploy the resulting neural models onto the Akida neuromorphic processor from BrainChip company [4]. The results obtained in terms of accuracy, latency and energy will be compared to other existing embedded solutions for Edge AI [2].
Project mission
The project mission will be organized in several periods:

• Bibliographic study on spiking neural network training
• Introduction to the existing Sw framework from BrainChip
• Training of convolutional neural networks for embedded applications [1] and conversion from CNN to SNN from Keras
• Deployment of the SNN onto Akida processing platform
• Experiments and measurements
• Publication in an international conference.

References
[1] L Cordone, Miramond B, Thierion, Object Detection with Spiking Neural Networks on Automotive Event Data, IEEE International Joint Conference on Neural Networks (IJCNN), 2022[2] N Abderrahmane, Miramond B, Kervennic E, A Girard, SPLEAT: SPiking Low-power Event-based
ArchiTecture
for in-orbit processing of satellite imagery, IEEE International Joint Conference on Neural Networks, 1-10, 2022

[3] E Lemaire, L Cordone, A Castagnetti, PE Novac, J Courtois, B Miramond, An Analytical Estimation of Spiking
Neural Networks Energy Efficiency, Springer International Conference on Neural Information Processing, 2022[4] T. Álvarez-Sánchez, et al, Detection of facial emotions using neuromorphic computation, Applications of
Digital Image Processing, 2022

Practical information
Location : LEAT Lab / SophiaTech Campus, Sophia Antipolis
Duration : 6 months from march 2023
Grant : from ANR project DeepSee
Profile : Machine learning, Artificial intelligence, Artificial neural networks, Python, Keras, Pytorch
Research keywords : Spiking neural network, Edge AI, neuromorphic computing

Contact and supervision
Benoît Miramond, Andrea Castagnetti
LEAT Lab – University Cote d'Azur / CNRS
Polytech Nice Sophia
04.89.15.44.39. / benoit.miramond@univ-cotedazur.fr

Le LEAT – LEAT

leat.univ-cotedazur.fr

THE LEAT​

• Home/
• The LEAT

Presentation​

The Laboratory of Electronics, Antennas and Telecommunications (LEAT) is a Joint Unit Université Côte d'Azur – CNRS (UMR n°7248). It is located on the SophiaTech campus, which is a training and research center dedicated to Information and Communication Technologies (ICT) involving academic actors (UNS, INRIA, EURECOM, CNRS, Polytech'Nice Sophia, Mines Paris Tech, etc.), competitiveness clusters, numerous associations and technological platforms.

Management Team​

Director: Robert Staraj (PR, UCA)
Deputy Directors: Jean-Marc Ribero (PR, UCA) and François Verdier (PR, UCA)
Administrative Manager: Françoise Trucas (CNRS)

Research activities​

Research activities are carried out in the field of telecommunications, radar, e-health, security, smart buildings, earth observation, sustainable development, etc. They are organized into three themes: EDGE (Edge Computing and Digital Systems), CMA (Antenna Design and Modeling) and ISA (Microwave Imaging and Antenna Systems).

Environment​

LEAT participates in the activities of the main competitiveness clusters in the region. It is associated with the research program of the "Laboratory of Excellence" Labex UCN@Sophia. Among the partners with which the laboratory is associated, are present the two supervisory bodies (Nice Sophia Antipolis University and the CNRS), but also Eurecom, Inria and the I3S and LTCI laboratories. The Labex is part of the IDEX UCAJEDI.
The Centre de RErecherche Mutualisé sur les ANTennes (CREMANT), created in 2008, is a joint laboratory between the University of Nice Sophia Antipolis (now Université Côte d'Azur), the CNRS and Orange. It allowed the pooling of personnel and equipment between LEAT academic researchers and Orange Labs La Turbie engineers on common research topics (antenna integration, engineering for e-health, multisensor, MIMO and massive MIMO systems, antennas based on new materials, electromagnetic modeling).

Little more info on DeepSee....might notice a particular partner to the project you might have heard of?

https://explore.openaire.eu/search/project?projectId=anr_________::3dea75a2c00248c04b90d2a1cf1e57fa

https://explore.openaire.eu/search/organization?organizationId=pending_org_::7207237ff84d2f09835522cbe5cfe575

 

[Violin1]

Anyone recall the competition here that someone was running to giveaway BRN shares?

@Risefromtheashes
Rise - are you talking about the legendary @butcherano who ran the footy tipping comp last year?

 

[Fact Finder]

Firstly if others want to go back and check I warned against the euphoria that some greeted the admission of Brainchip into the ASX200. (I warned all at the time it would bring to the company a new level of short activity. I suggested this would not be good for the share price and short term investors.)

And it did.

Early in the HOT Ukraine war I warned that this war was going to crush economic activity in the EU, UK and as a result hurt the US at a time when it was flexing its muscles in a trade war with China.

And it did.

I also stated that all these Global storms would however work to Australia’s advantage. I said China needed Australia more than Australia needed China. I said that Europe & the UK needed Australia to fill in the energy gap left by Russia destroying for decades to come any thoughts that Europeans and British peoples had that Russia could be a reliable trading partner. I also pointed to the unsavoury fact that Brainchip would benefit from the resultant increase in military spending by the US and its allies.

And it did.

I said many times the lucky country would continue to prove it was the Economic Miracle.

And it did.

While Brainchip is an Australian born company like Atlassian it too is not a company that finds its success in Australia. As the CEO Sean Hehir said at the AGM shareholders need ‘to get their heads out of their backsides’ and accept that Brainchip is an international company. (He did not use ‘their heads out of their backsides’ line but this was his intent.) Brainchip’s future is international and as such it is subject to all the economic and trading headwinds that circulate in the international economy. It cannot shelter from these forces by hiding away in the economic miracle and become a technology giant. It has to go international.

And it did.

Against this background of increased short activity by ASX200 practitioners including activist shorts, global supply chain disruption, trade wars involving Russia, China, USA and Western countries Brainchip warned of a slow down in the markets it inhabits. Despite these factors Brainchip pushes to grow its ecosystem and increase its sales and marketing efforts across the globe.

And it did.

As for sales cycles I pointed firstly to Prophesee and Sony and it’s shortly to be released mass market vision sensor. Why because it is a perfect example of a neuromorphic product being sold to and developed by a non neuromorphic global company. The evidence is clear Prophesee and Sony commenced working together in 2020. It is now 2023. Time to market at least three years. Despite this great engagement Prophesee raised $50 million in capital late last year to keep the doors open. Substantial income clearly unlikely to start to flow from Sony to Prophesee before end 2023.

But it will.

Renesas announced by Brainchip as an IP purchaser 23.12.20. Renesas announces tape out of AKIDA powered chip just recently with mass production late 2023. Income will flow in 2024.

And it will.

Socionext VP of Sales committed to assisting Brainchip to commercialise the AKIDA product family and just a month or so before the full commercial release of the AKD1000 in 2021 Socionext commenced to promote AKD1000 at trade fairs using the AKD1000 engineering samples as demonstrators. In January, 2023 Socionext announced at CES that they have a range of AKIDA based automotive solutions for customer consideration.

And they are.

Mercedes Benz revealed they were working with Brainchip at CES2022 with vehicles coming to market in 2024 and Brainchip have stated they continue to work with Mercedes Benz to achieve this end.

And they will.

Nviso struggled to find traction in the market place for 11 years and now having discovered Brainchip’s AKIDA solution is finally able to realise its ambition.

And it will.

Prophesee was building a glass house until it was completed by Brainchip’s AKIDA after having first contact with Adam Osseiran 6 years ago is looking towards a commercial engagement.

And it will.

Intel, ARM & SiFive having been announced over 2022 are engaged in marketing Brainchip’s AKIDA solution along with Edge Impulse.

And they are.

MegaChips having first discovered Brainchip’s AKIDA technology solution in 2020 set about creating a section within its organisation of engineers trained in and dedicated to the use and implementation of the AKIDA technology solution and in late 2021 they with Brainchip announced the purchase of a full AKIDA IP licence and set about marketing AKIDA based solutions to their customer base.

And they do.

So just like the Little Red Hen who found some grain and wanted to make bread despite the criticism of the other animals in the barnyard who are lazy or ignorant and not prepared to suffer the hard work and delays involved in taking wheat through the process necessary to become bread Brainchip just keeps forging ahead in an industry where from first customer contact to product on shelf is three years plus and where paradigm shifting is slightly more difficult than alchemy.

My opinion only DYOR
FF

AKIDA BALLISTA

Apologies to the classically trained but I am reading to an audience of under 2 year olds where The Little Red Hen rules.

Now that a few of you have had time to read the above post and noticed the reference to Atlassian I recommend the following:
My opinion only DYOR
FF

AKIDA BALLISTA

Our Q1 FY23 letter to shareholders​

PUBLISHED NOVEMBER 3, 2022 IN COMPANY NEWS

SCOTT FARQUHAR
Co-Founder and Co-CEO

MIKE CANNON-BROOKES
Co-Founder and Co-CEO

CAMERON DEATSCH
Chief Revenue Officer

JOE BINZ
Chief Financial Officer
An update to customers, stakeholders, and shareholders on our mission to unleash the potential in every team.

prefer a PDF?​

No worries! Grab it here.
Fellow shareholders,
Atlassian’s mission to unleash the potential of every team advances ever onward, and Q1 was another momentous quarter of executing against our long-term initiatives. We announced a new subscription offering, launched a product into general availability, and held our first large-scale customer event focused on a single market, which got rave reviews from attendees. That said, companies in nearly every industry are facing headwinds, and we’re beginning to see the impact on our business.
In the spirit of our “Open company, no bullshit” value

, let’s start with the topic that’s top of mind for shareholders: macroeconomic impacts.

1. Last quarter, we shared that we saw a decrease in the rate of Free instances converting to paid plans. That trend became more pronounced in Q1.
2. This quarter, we started to see a slowing in the rate of paid user growth from existing customers.

To be clear, we’re not seeing any changes in our competitive position or in the inherent demand for our products. Looking across our customer base of 249,000+, there has been no overall decrease in usage or change in churn. The above two trends are the result of companies tightening their belts and slowing their pace of hiring. In other words, Atlassian is not immune to broader macro impacts. Our outlook assumes these trends will persist, but we’ll monitor, respond, and keep you updated accordingly.
Turbulent markets provide an opportunity to shake up the leaderboards, and we are poised to play offense in this environment. Buoyed by the secular tailwinds of digital and cultural transformation, Atlassian is incredibly well-positioned to capture additional share in each of our three massive markets – agile/DevOps, IT service management, and work management – and we’re working to do just that. In particular, we have huge opportunities in cloud migrations, serving enterprises, and ITSM – areas where we’ve already seen significant momentum and strong ROI. In the past year, migrations and enterprise deals in the cloud were both up more than 2x, and Jira Service Management added 10k customers.
We will focus our investments on strengthening our market position and scooping up top-tier talent in this environment. But we will balance these investments with the growth of our business and be responsive to macroeconomic conditions. So while we’re lowering our revenue outlook for FY23 based on macroeconomic headwinds, we are maintaining our mid-teens % operating margin outlook for the year. (For further detail, see the Fiscal 2023 Outlook section below.)
Despite the near-term instability in the world around us, we remain certain about the incredible long-term opportunities in front of Atlassian and our ability to capitalize on them. We’ve talked about having a line of sight to $10B in annual revenue. This hasn’t changed.
We have the right products, the right leaders, and the right strategies in place to come out of this downturn in an even stronger position. When obstacles emerge along the way, we’ll navigate around them as we always have: with vigilance, pragmatism, and agility.

 

[BaconLover]

I don't understand the hatred towards Tesla, must be an ego thing.

Tesla has done brilliant things, proof and evidence is on their balance sheets.
SpaceX is amazing and Twitter has never been better.
We cannot see the same success on Brainchip balance sheets yet, that's a fact.
I'm a firm supporter of Brainchip and wants it to be successful but when comparing to Tesla, who's the real show pony?

Sean is a clever CEO in my opinion.

LDN said 2021 will be the year of Brainchip.
Peter said 2022 Dec we will see things explode.

Sean is clever - he said watch the financials, but didn't say which year
Clever indeed.
Those who hid under the rocks when things were falling apart are now returning to tell other share holders how to dance to their tune.
Good luck with that.

 

[Diogenese]

Looks like this Team want to explore us a little further

Haven't looked into who or what ANR Project DeepSee is yet - that's where the grant is coming from

Evaluation of neuromorphic AI with embedded Spiking Neural Networks – LEAT

• DECEMBER 10, 2022
• COMMENTS OFF

Evaluation of neuromorphic AI with embedded Spiking Neural Networks​

Background
AI is proliferating everywhere even to embedded systems to integrate intelligence closer to the sensors (IoT, drones, vehicles, satellites ...). But the energy consumption of current Deep learning solutions makes classical AI hardly compatible with energy and resource constrained devices. Edge AI is a recent subject of research that needs to take into account the cost of the neural models both during the training and during the prediction. An original and promising solution to face these constraints is to merge compression technics of deep neural networks and event-based encoding of information thanks to Spiking neural networks (SNN). SNN are considered as third generation of artificial neural networks and are inspired from the way the information is encoded in the brain, and previous works tend to
conclude that SNN are more efficient than classical deep networks [3]. This internship project aims at confirming this assumption by converting classical CNN to SNN from standard Machine Learning frameworks (Keras) and deploy the resulting neural models onto the Akida neuromorphic processor from BrainChip company [4]. The results obtained in terms of accuracy, latency and energy will be compared to other existing embedded solutions for Edge AI [2].
Project mission
The project mission will be organized in several periods:

• Bibliographic study on spiking neural network training
• Introduction to the existing Sw framework from BrainChip
• Training of convolutional neural networks for embedded applications [1] and conversion from CNN to SNN from Keras
• Deployment of the SNN onto Akida processing platform
• Experiments and measurements
• Publication in an international conference.

References
[1] L Cordone, Miramond B, Thierion, Object Detection with Spiking Neural Networks on Automotive Event Data, IEEE International Joint Conference on Neural Networks (IJCNN), 2022[2] N Abderrahmane, Miramond B, Kervennic E, A Girard, SPLEAT: SPiking Low-power Event-based
ArchiTecture
for in-orbit processing of satellite imagery, IEEE International Joint Conference on Neural Networks, 1-10, 2022

[3] E Lemaire, L Cordone, A Castagnetti, PE Novac, J Courtois, B Miramond, An Analytical Estimation of Spiking
Neural Networks Energy Efficiency, Springer International Conference on Neural Information Processing, 2022[4] T. Álvarez-Sánchez, et al, Detection of facial emotions using neuromorphic computation, Applications of
Digital Image Processing, 2022

Practical information
Location : LEAT Lab / SophiaTech Campus, Sophia Antipolis
Duration : 6 months from march 2023
Grant : from ANR project DeepSee
Profile : Machine learning, Artificial intelligence, Artificial neural networks, Python, Keras, Pytorch
Research keywords : Spiking neural network, Edge AI, neuromorphic computing

Contact and supervision
Benoît Miramond, Andrea Castagnetti
LEAT Lab – University Cote d'Azur / CNRS
Polytech Nice Sophia
04.89.15.44.39. / benoit.miramond@univ-cotedazur.fr

Le LEAT – LEAT

leat.univ-cotedazur.fr

THE LEAT​

• Home/
• The LEAT

Presentation​

The Laboratory of Electronics, Antennas and Telecommunications (LEAT) is a Joint Unit Université Côte d'Azur – CNRS (UMR n°7248). It is located on the SophiaTech campus, which is a training and research center dedicated to Information and Communication Technologies (ICT) involving academic actors (UNS, INRIA, EURECOM, CNRS, Polytech'Nice Sophia, Mines Paris Tech, etc.), competitiveness clusters, numerous associations and technological platforms.

Management Team​

Director: Robert Staraj (PR, UCA)
Deputy Directors: Jean-Marc Ribero (PR, UCA) and François Verdier (PR, UCA)
Administrative Manager: Françoise Trucas (CNRS)

Research activities​

Research activities are carried out in the field of telecommunications, radar, e-health, security, smart buildings, earth observation, sustainable development, etc. They are organized into three themes: EDGE (Edge Computing and Digital Systems), CMA (Antenna Design and Modeling) and ISA (Microwave Imaging and Antenna Systems).

Environment​

LEAT participates in the activities of the main competitiveness clusters in the region. It is associated with the research program of the "Laboratory of Excellence" Labex UCN@Sophia. Among the partners with which the laboratory is associated, are present the two supervisory bodies (Nice Sophia Antipolis University and the CNRS), but also Eurecom, Inria and the I3S and LTCI laboratories. The Labex is part of the IDEX UCAJEDI.
The Centre de RErecherche Mutualisé sur les ANTennes (CREMANT), created in 2008, is a joint laboratory between the University of Nice Sophia Antipolis (now Université Côte d'Azur), the CNRS and Orange. It allowed the pooling of personnel and equipment between LEAT academic researchers and Orange Labs La Turbie engineers on common research topics (antenna integration, engineering for e-health, multisensor, MIMO and massive MIMO systems, antennas based on new materials, electromagnetic modeling).

LEAT Lab – University Cote d’Azur ...
tough work ... but someone's gotta do it!

 

[db1969oz]

I don't understand the hatred towards Tesla, must be an ego thing.

Tesla has done brilliant things, proof and evidence is on their balance sheets.
SpaceX is amazing and Twitter has never been better.
We cannot see the same success on Brainchip balance sheets yet, that's a fact.
I'm a firm supporter of Brainchip and wants it to be successful but when comparing to Tesla, who's the real show pony?

Sean is a clever CEO in my opinion.

LDN said 2021 will be the year of Brainchip.
Peter said 2022 Dec we will see things explode.

Sean is clever - he said watch the financials, but didn't say which year
Clever indeed.
Those who hid under the rocks when things were falling apart are now returning to tell other share holders how to dance to their tune.
Good luck with that.

 

[Diogenese]

Part of the same research by the looks. Oct 22.

https://arxiv.org/pdf/2210.13107.pdf

https://leat.univ-cotedazur.fr/2022...d-object-detection-with-bio-inspired-retinas/

Development of a prototype HW platform for embedded object detection with bio-inspired retinas​

Context
The LEAT lab is leader of the national ANR project DeepSee in collaboration with Renault, Prophesee and 2 other labs in neuroscience (CERCO) and computer science (I3S). This project aims at exploring a bio-inspired approach to develop energy-efficient solutions for image processing in automotive applications (ADAS) as explored by [3]. The main mechanisms that are used to follow this approach are event-based cameras (EBC are considered as artificial retinas) and spiking neural networks (SNN).
The first one is a type of sensor detecting the change of luminosity at very high temporal resolution and low power consumption, the second one is a type of artificial neural network mimicking the way the information is encoded in the brain. The LEAT has developed the first model of SNN able to make object detection on event-based data [1] and the related hardware accelerator on FPGA [2]. The goal of this internship project is to deploy this spike-based AI solution onto an embedded smart camera provided by the Prophesee company [4]. The camera is composed of an event-based sensor and an FPGA. The work will mainly consist in deploying the existing software code (in C) on the embedded CPU, integrate the HW accelerator (VHDL) onto the FPGA and make the communication between them through an AXI-STREAM bus. The last part of the project will consist in realizing experimentations of the resulting smart cameras to evaluate the real-time performances and energy consumption before a validation onto a driving vehicle.