BRN Discussion Ongoing

[Esq.111]

Evening Chippers ,

Well bend me over & spank me with a feather duster...

We are pressently up 10.83% on the German market , under two hours in , looking promising .
21 thousand € traded so far .

Regards,
Esq

 

[Frangipani]

IMEC has caught the European MemRistor disease:

EP3671748A1 IN-MEMORY COMPUTING FOR MACHINE LEARNING 20181221

View attachment 50727


EP3671750A1 SYNAPSE CIRCUIT WITH MEMORY

View attachment 50726



US11443174B2 Machine learning accelerator
View attachment 50725

Hi @Diogenese

while I don’t have the background to spar with you on technical matters, I noticed that you posted links to three different patents all filed back in 2018 (!) by a couple of researchers based at the imec headquarters in Leuven, Belgium , whereas the team currently working on the RISC-V-based SENeCA digital neuromorphic processor that @Bravo referred to is actually from imec The Netherlands in Eindhoven, which also happens to be the birth place of Philips (hence it is a natural fit for them to collaborate, I’d say - European localism should not be underestimated).

Ella and I both reckon that just because the researchers named in those older patents you linked and those working on SENeCA as I write are all imec employees, it doesn’t mean those independent teams must have a similar approach?
Or put differently: Unlike Ella and Louis, they may not necessarily be dancing Cheek to Cheek… 🎙

Cheers,
Frangipani

 

[Baron Von Ricta]

Perhaps everyone is busy listening to the new podcast
Good day today chippers

The Big 'L'

I have discovered why the share price rose today.
It has been predicted by Vanga – known as the “Nostradamus of the Balkans” – was believed to have had an extraordinary ability to predict the future.

Baba Vanga’s 2024 predictions: from Putin’s assassination plot to rise of terror attacks and medical breakthroughs​

6. Tech revolutions

The clairvoyant had a vision that there will be a major breakthrough in quantum computing,

It comes as quantum computing is rapidly progressing and can solve issues faster than traditional computers.

If the prediction turns out to be true, 2024 could see the power of AI skyrocketing in society.

7. Medical breakthroughs

Last but most certainly not least, Baba has offered what could be some of the best in the medical field.

The blind Bulgarian claimed there will be new treatments for incurable diseases including Alzheimer’s and cancer in 2024.

So there you have it. I'm excited.

 

[Perhaps]

Can someone please call Rob Telson ASAP to get him to speak to someone at Phillips Medical Systems?

I just stumbled upon this Power Point Preso which shows Phillips are working with IMEC on the denoising of medical XRAYs using SNN's (SENECA) and they're hoping to apply this same technology can be applied in WIDE RANGE OF CLINICAL APPLICATION AREAS.

I believe that SENECA is still only at research stage which could leave the door wide open for us to say hello to Phillips. It says in this research paper below, published 23 June 23 that " The SENECA platform and the tools used in this project are available for academic research upon request."

Then there's also this Linkedin response from 5 months ago where the researcher at IMEC says they can't compare SENECA with BrainChip which he differentiates as being "commercial neuromorphic architecture".

SENECA: building a fully digital neuromorphic processor, design trade-offs and challenges

Neuromorphic processors aim to emulate the biological principles of the brain to achieve high efficiency with low power consumption. However, the lack of flexibility in most neuromorphic architecture designs results in significant performance loss and ...

www.ncbi.nlm.nih.gov

https://tempo-ecsel.eu/wp-content/uploads/2022/03/TEMPO-poster-demo-image_denoising.pdf

View attachment 50712
View attachment 50713

This is all part of the European Tempo project/Andante consortium. They are in R&D since 2018 and go their own way, based on SynSense technology from the beginning.
This is all about developing European solutions in the neuromorphic field, so the use of US technology just doesn't fit the bill.
It's better to file this under possible future competition.
Sidenote:
Mercedes was also a part of the Tempo project in the beginning, but stepped out in 2019, possibly in favour of using Brainchip technology.
Imec, Philips and Valeo are driving forces of this powerful consortium since 2018, led by STMicroelectronics.
A political view to think about:
Brainchip offers a new AI technology which is under heavy development by NASA/DARPA. When it comes to a heavy use in the military complex, are there any guarantees there won't be US restrictions for export?

AI vision chip for autonomous driving reaches tape-out stage

Deep learning and image processing expert Videantis GmbH has reached the tape-out stage with its low-power AI chip. The development is architecturally based on the v-MP6000UDX unified processing platform and targets various autonomous driving use cases.

www.eenewseurope.com

 

[Fullmoonfever]

Nice to see MBRDI get engineers onboard with neuromorphic skills

Rushikesh Patil​

PGET in Mercedes Benz Research and Developement India​

Mercedes-Benz Research and Development India Indian Institute of Space Science and Technology​

Thiruvananthapuram, Kerala, India
990 followers 500+ connections

​

About​

I have strong background in mathematics and skilled in Neuromorphic circuits, ‌‌analog, mixed signal circuit , digital vlsi circuit designing, deep learning,machine learning, Verilog

Experience​

• 
  

  Engineer​

  Mercedes-Benz Research and Development India​

  Aug 2023 - Present 4 months
  Bengaluru, Karnataka, India
  HiL, FPGA,Modelling in Simulink, Python Automation, Embedded System, Microcontroller, System Verilog
• 
  

  Project Intern​

  Mercedes-Benz Research and Development India​

  Aug 2022 - Jan 2023 6 months
  Bengaluru, Karnataka, India
  Testing Quality Automation and OBD Automation using Python and VB.NET

 

[MDhere]

Evening Chippers ,

Well bend me over & spank me with a feather duster...

We are pressently up 10.83% on the German market , under two hours in , looking promising .
21 thousand € traded so far .

Regards,
Esq

Be careful what u preach rocket will have a gif for u on that lol

 

[JB49]

This may have already been posted here. But potentially a few dots.

- A thales staff member working on neuromorphic computing https://www.thalesgroup.com/en/worldwide/group/news/julie-grollier-nano-neuron-architect
- Julie grollier wrote this paper about neuromorphic computing and Spiking Neural networks https://www.sciencedirect.com/science/article/pii/S2589004221001905
- Thales is working with Western sydney university who developed core intellectual property for an event based neuromorphic sensor, astrosite. Could this be why Western Sydney University have been tight-lipped about any links with us? https://www.thalesgroup.com/en/coun...ace-capability-australia-international-effort

 

[Fastback6666]

Don't believe it. They're having file loading problems again. If you can go back and see daily trades, 99.5% of trades at the centrepoint ie 18.25c or 18.75c etc is a short

Shortman reporting shorts jumped up 5m on the 22nd.

View attachment 50733

Snapshot of some of the trades today on the Chi-X

2:34:07 pm	2:42:23 pm	21.75	104,395	0.25	$22,705	12
2:33:24 pm	2:33:24 pm	21.5	10,555	0.25	$2,269	1
2:32:15 pm	2:32:15 pm	21.75	1,446	0.25	$314	1
2:32:09 pm	2:32:09 pm	21.5	58	0.25	$12	1
2:21:58 pm	2:29:58 pm	21.75	14,599	0.25	$3,175	7
2:21:08 pm	2:21:08 pm	21.5	2,857	0.25	$614	1
2:16:08 pm	2:16:10 pm	21.75	75,500	0.25	$16,421	3
2:11:10 pm	2:16:08 pm	21.5	494,700	0.25	$106,360	13
2:01:45 pm	2:01:45 pm	21.75	6,797	0.25	$1,478	1
2:01:43 pm	2:01:43 pm	21.5	287,093	0.25	$61,724	19
1:57:47 pm	2:01:43 pm	21.25	50,000	0.25	$10,625	3
1:52:21 pm	1:52:21 pm	21.5	6,373	0.5	$1,370	2
1:51:41 pm	1:52:18 pm	21	231,561	0.5	$48,627	10
1:45:58 pm	1:45:58 pm	21.5	6,564	0.5	$1,411	1
1:41:54 pm	1:45:58 pm	21	186,637	0.25	$39,193	11
1:39:08 pm	1:41:54 pm	21.25	37,003	0.25	$7,863	4
1:29:37 pm	1:38:45 pm	21.5	154,701	0.25	$33,260	22
1:26:58 pm	1:29:07 pm	21.75	21,292	0.25	$4,631	2
1:26:37 pm	1:26:37 pm	21.5	10,000	0.25	$2,150	1
1:19:01 pm	1:25:58 pm	21.75	26,691	0.25	$5,805	3
1:17:56 pm	1:17:56 pm	21.5	25,876	0.25	$5,563	3
1:08:03 pm	1:17:14 pm	21.75	39,355	0.25	$8,559	4
1:08:03 pm	1:08:03 pm	21.5	51,000	0.25	$10,965	3

Well the shorts are very brave to take any new positions at 20 cent area.

One decent bit of news that relates to revenue and we will be back to 40 cents +++

On todays trading;
Well we have to be happy with a near 13% day, if it can crack through 25 cents we could be on a new upward trend. News is key though.

 

[Boab]

Well the shorts are very brave to take any new positions at 20 cent area.

One decent bit of news that relates to revenue and we will be back to 40 cents +++

On todays trading;
Well we have to be happy with a near 13% day, if it can crack through 25 cents we could be on a new upward trend. News is key though.

Lets hope that the shorters are now thinking "this thing has had a good run, now it's time to move on to something else".

 

[Diogenese]

Both Sally and the other presenter admitted they do not understand Akida. It looks like we need a media pack which explains Akida's spiking neuromorphic processing unit.

Akida originally used a single binary bit (1-bit) "spike".

This was field tested (FPGA?) with EAPs and was changed to an upper limit of 4-bit, with 1-bit and 2-bit options for lower power. This change meant that the spike could assume any value from 1 to 15, greatly increasing accuracy/sensitivity. These multi-bit options would have required an increase to 4 in the number of conductors in the SoC copper bus so the bits could be transmitted in parallel. Likewise, the processing would have become more complex, requiring the capability to handle 4-bit processes.

Akida 2 now has the 8-bit option. From the little I've been able to glean, I believe Akida 2 only uses the 8-bit option on the input stage, reverting to 4-bits for the internal layers.

It would be an hour's work for an Akida engineer to knock together a few simplified functional block diagrams illustrating the 1-bit, 4-bit, and 8-bit versions of Akida.

Just to add:

neuromorphic computing is a step beyond CPU/GPU.

SNN is a step beyond neuromorphic.

CPU/GPU are based on mathematical precision. Hence floating point numbers (usually 32 bits or more with exponential multipliers) which is what MACs (Multiply Accumulate Circuits) are very good at. A MAC multiplies one number (eg 32 bits) by another 32 bit number - that's about 1000 operations, and then adds (accumulates) the result with previous results. For an image, this process is carried out for each pixel in an image in a series of comparisons. This amounts to billions of operations (transistor switching on/off which takes time and draws electrical power).

The brain does not use mathematical precision to interpret the signals from the 5 senses. Instead it compares the signals from its sensors with memories of previous experiences/previously experienced sensor signals.

But this comparison does not use mathematical precision. Instead, it uses the nearest approximation - "this smells more like toast than cake".

Using images as an example, a SNN takes small samples of the input image and compares them with small samples from the model library of images. from PvdM's 4 Bis Are Enough paper, we can infer that the output of each pixel can be adequately represented with sufficient accuracy by 4 bits.

An SNN does not produce mathematical certainty. It produces the highest probability.

Hi @Diogenese

while I don’t have the background to spar with you on technical matters, I noticed that you posted links to three different patents all filed back in 2018 (!) by a couple of researchers based at the imec headquarters in Leuven, Belgium , whereas the team currently working on the RISC-V-based SENeCA digital neuromorphic processor that @Bravo referred to is actually from imec The Netherlands in Eindhoven, which also happens to be the birth place of Philips (hence it is a natural fit for them to collaborate, I’d say - European localism should not be underestimated).

Ella and I both reckon that just because the researchers named in those older patents you linked and those working on SENeCA as I write are all imec employees, it doesn’t mean those independent teams must have a similar approach?
Or put differently: Unlike Ella and Louis, they may not necessarily be dancing Cheek to Cheek… 🎙

Cheers,
Frangipani

View attachment 50742

View attachment 50742






View attachment 50744


View attachment 50746


View attachment 50747

Hi Frangipani,

One again you've tripped me up with your geographic prestidigitations.

And then you throw this ensemble at me:
https://en.wikipedia.org/wiki/Ella_and_Louis#The_album - just look at that band! Oscar Peterson, Herb Ellis, Ray Brown, Buddy Rich!
Peterson's Easy Listenin Blues is the epitome of the smoky 3 AM jazz dive (leaving aside Hawkins Smoke gets in Your Eyes).
Herb Ellis is still on the town, but Ray Brown is reportedly behavin' himself, and Buddy Rich beat it.



Yes, those pesky Belgian IMECers do have a digital CNN accelerator patent:

EP3674982A1 HARDWARE ACCELERATOR ARCHITECTURE FOR CONVOLUTIONAL NEURAL NETWORK 20181227

A hardware accelerator architecture (10) for a convolutional neural network comprises a first memory (11) for storing NxM activation inputs of an input tensor; a plurality of processor units (12) each comprising a plurality of Multiply ACcumulate (MAC) arrays (13) and a filter weights memory (14) associated with and common to the plurality of MAC arrays of one processor unit (12). Each MAC array is adapted for receiving a predetermined fraction (FxF) of the NxM activation inputs from the first memory, and filter weights from the associated filter weights memory (14). Each MAC array is adapted for subsequently, during different cycles, computing and storing different partial sums, while reusing the received filter weights, such that every MAC array computes multiple parts of columns of an output tensor, multiplexed in time. Each MAC array further comprises a plurality of accumulators (18) for making a plurality of full sums from the partial sums made at subsequent cycles.

... but it has the Scottish disease.

SENECA unfortunately suffers from hysteron proteron (ACENES) - it uses the processor system to run the NPEs:

3 ... In general, for a normal incoming spike, RISC-V performs a pre-processing phase to retrieve the relevant local information required to process the spikes (for example, the address of the corresponding parameters in the Data Memory) and packs that information in the form of a micro-task. Then this micro-task is pushed to the Task FIFO. The loop controller executes the tasks one by one based on the micro-code instructions stored in the loop buffer. The loop controller is a small dedicated controller programmed to execute a sequence of instructions in parallel through the NPEs (Neural Processing Elements). Some neural operations in NPEs may result in output spikes which will be converted to packets of data inside the event generator. The event generator unit interrupts the RISC-V to perform postprocessing on the generated events. RISC-V can feed the generated events back into the Task FIFO or send them out through the NoC. Following, we will explain each element of SENECA in more detail.


I don't know how Akida will interact with RISC-V, but it will not tolerate interruptions from the system.

3.2. Neuron processing elements (NPEs)

The SENECA core includes an array of neuron processing elements (NPEs) that act as physical neurons in Figure 1C. Each NPE contains a small register-based memory and executes a category of instructions. An array of NPEs is forming a SIMD (Single Instruction Multiple Data) type architecture (Flynn, 1972). Instructions to be executed in NPEs are coming from the Loop Buffer. NPEs can get their data from Data Memory (through a wide Data Memory port), RISC-V (by directly writing into their register file), and Loop controller (broadcasting).

The register file inside the NPEs allows for reusing data as much as possible before reading/writing it into the Data Memory. Table 2 shows that accessing the data in NPEs' register file is about 20 × more energy efficient than accessing the Data in the Data Memory (SRAM). For example, in an extreme case where the number of neurons is low6, keeping the neuron states inside the NPEs and only reading the weights from Data Memory (avoiding the neuron state read/write) reduces the energy consumption of a synaptic operation from 2.8 to 1.8 pJ7.

In neuromorphic applications, the optimized resolution of neuron states and synaptic weights depends on several variables (Khoram and Li, 2018). Therefore, to optimize the memory footprint and access energy, it is crucial that our NPEs support various data types and precision. Currently, NPEs are designed to support 4, 8, and 16 bit data precisions, both for linear and logarithmic quantization (floating point). They also support shared scale factors (Köster et al., 2017; Moons et al., 2017; Jacob et al., 2018; Kalamkar et al., 2019; Coelho et al., 2021). This flexibility allows for the memory-efficient deployment of mixed precision neural networks for inference and on-device adaptation. Each NPE consumes 1.3% of the total area of the core.

3.3
...
As mentioned, NPEs do not implement a specific neuron model. They only execute special operations, which are common among many neuron models. A neuron/synapse/learning model can be built by sequential execution of a few instructions, called microcode. The loop controller sends the microcode to the NPEs in a “for-loop” style to process events. Therefore, the Loop controller is optimized to execute nested loops. Executing loops using the loop controller is 100 × more energy efficient compared to the RISC-V.


SENECA is not a digital SNN SoC. It uses special programmable processors to perform SNN functions under software control.

It's not really surprising that they did not compare SENECA with Akida.

 

[Proga]

Well the shorts are very brave to take any new positions at 20 cent area.

One decent bit of news that relates to revenue and we will be back to 40 cents +++

On todays trading;
Well we have to be happy with a near 13% day, if it can crack through 25 cents we could be on a new upward trend. News is key though.

On no news, they'd be happy to sell into the higher price/volume and hope like the other 99% of times the price drifts back down. Easy money.

 

[Rach2512]

Who are the leaders in neuromorphic computing?

Who are the leaders in neuromorphic computing? TS2 SPACE

ts2.space

Sorry, not sure if previously posted, from October

Snippet

Additionally, BrainChip Holdings, an innovative technology company, has made noteworthy contributions to neuromorphic computing. Their Akida™ Neuromorphic System-on-Chip (NSoC) is specifically designed to process data in a manner similar to the human brain. With its low power consumption and high-speed processing capabilities, BrainChip’s Akida™ NSoC is well-suited for edge computing applications, including surveillance, cybersecurity, and more. BrainChip’s dedication to developing efficient and powerful neuromorphic computing solutions has solidified their position as a key leader in this domain.

 

[M_C]

Sign Up | LinkedIn

500 million+ members | Manage your professional identity. Build and engage with your professional network. Access knowledge, insights and opportunities.

www.linkedin.com

 

[cosors]

Evening Chippers ,

Well bend me over & spank me with a feather duster...

We are pressently up 10.83% on the German market , under two hours in , looking promising .
21 thousand € traded so far .

Regards,
Esq

averagely little

 

[KMuzza]

CSIRO study identifies AI models to improve automated chest X-ray diagnoses

The research from CSIRO's Australian e-Health Research Centre demonstrates the potential for AI to better support clinicians.

www.csiro.au

AKIDA BALLISTA UBQTS

 

[Schnitzel lover]

Up almost 13% today and nary a whisper from the forum..
What's going on ..

No green babies?..

And where's @Schnitzel lover? Surely he would have sold for a healthy profit today and be bagging the Company, to get back in again lower?..

Heyyy wait a minute..

View attachment 50732

Focusing on enjoying the green day yesterday Dingo..

 

[Schnitzel lover]

I’m not trading it @DingoBorat until it clears the 20Weekly..

 

[buena suerte :-)]

Who are the leaders in neuromorphic computing?

Who are the leaders in neuromorphic computing? TS2 SPACE

ts2.space

Sorry, not sure if previously posted, from October

Snippet

Additionally, BrainChip Holdings, an innovative technology company, has made noteworthy contributions to neuromorphic computing. Their Akida™ Neuromorphic System-on-Chip (NSoC) is specifically designed to process data in a manner similar to the human brain. With its low power consumption and high-speed processing capabilities, BrainChip’s Akida™ NSoC is well-suited for edge computing applications, including surveillance, cybersecurity, and more. BrainChip’s dedication to developing efficient and powerful neuromorphic computing solutions has solidified their position as a key leader in this domain.

Great post ... It has been posted a couple of times Rach

 

[IloveLamp]

One Giant Superchip for LLMs, Recommenders, and GNNs: Introducing NVIDIA GH200 NVL32 | NVIDIA Technical Blog

At AWS re:Invent 2023, AWS and NVIDIA announced that AWS will be the first cloud provider to offer NVIDIA GH200 Grace Hopper Superchips interconnected with NVIDIA NVLink technology through NVIDIA DGX…

developer.nvidia.com

 

[IloveLamp]

News Announcement - NeuReality

NeuReality Launches NR1™ AI Inference Solution in Time to Unleash Generative AI Across All Industries State-of-the-art AI inference system bypasses bottlenecks in CPU-center data centers, driving higher performance with up to 90 percent savings for every dollar spent on AI deployments. (Denver...

neureality.ai

Arm on LinkedIn: #onarm

NeuReality is on a mission to make ML inference easier to deploy and access. That's why they've turned to our versatile and flexible Arm Neoverse platform for…

www.linkedin.com