BRN Discussion Ongoing

I’m still excited about what Akida1000 can do. Gen 2 with TENNs must be impressive. How good is this?


APPLICATION​

A single SNAP Card, deployed on an airborne ISR platform, occupying only one 3U VPX slot, could deploy up to five complex machine learning algorithms across each of its onboard processors to perform the following missions simultaneously.

– Automated Target Recognition (ATR) on full motion video feeds and imagery (to include 4k and higher)​

– Real-time detection and identification of threat radars and their acquisition/operating modes​

– Detection of FISINT (Foreign Instrumentation Signature Intelligence) and/or hacking across the airframe’s 1553 communications bus​

– Perform communications analysis to include speech-to-text and foreign language translation of intercepted communications​

– Fuse multiple infrared cameras (e.g., SWIR, MWIR, LWIR) to provide a combined infrared operating picture on the ground Multi-user, simultaneous modulation/demodulation​


1734517485770.png
 
  • Like
  • Love
  • Fire
Reactions: 59 users

Diogenese

Top 20
Mercedes seeking a PhD commencing Feb 2025!

50/50 chance being Loihi based on previous studies.

Project overview:

We are looking for a PhD student for our innovative project to develop an end-to-end autonomous neuromorphic driving system. Based on methods of neuromorphic computing (e.g. Spiking Neural Networks), new models are developed and their energy consumption and computing efficiency are tested in a simulator for a neuromorphic chip. The project is a collaboration between Mercedes-Benz AG and the University of Waterloo (Canada). The successful applicant will be contacted by Prof. Hendrik Lensch (University of Tübingen) and Prof. Chris Eliasmith (University of Waterloo). During the doctorate, several research stays at the University of Waterloo are planned for the doctoral student


Chris Eliasmith, who seems to be quite a polymath, is also CTO of ABR. Some of ABR's patents seem to have sprung from Canada's equivalent of CSIRO.

MB signed an MOU with Ontario in 2022 for cooperation across the electric vehicle chain. In addition to the Wellington Uni research centre linked to Eliasmith, MB has joined the Ontario start-up incubator OVIN.

ABR have patents for their own in-house NN. A sampling follows:

WO2024197396A1 EVENT-BASED NEURAL NETWORK PROCESSING SYSTEM 20230326

1734516150430.png



[0027] The disclosed processing system is highly flexible, in that individual processing elements performing the neural network evaluation can in principle be arranged and connected in any manner. For example, in the processing system may be customized for specific purposes by instantiating the individual components of the system in a hardware description language (e.g., SystemVerilog™) and can be modified by the person of ordinary skill in the art without requiring deep understanding of low-level implementation details. Furthermore, the individual components of the processing system place relatively few constraints on possible connectivity. For instance, the event bus system supports an arbitrary number of input ports by automatically distributing events through a balanced binary tree of bus arbiters, and phase synchronisation and pipelining happens automatically through stream connections. This enables the processing system to be scaled to field programmable gate arrays (FPGAs) of various sizes, and to be adapted to support specific neural network topologies. For example, a specific class of networks may require many neurons in the first layers, but only a few neurons in subsequent layers. The network topology can be adapted accordingly.

[0029] To support 2D convolution, the disclosed processing system tags events with 2D coordinates that describe what source location the event corresponds to. This source location determines what target neurons need to be updated
.

US2021342668A1 Methods And Systems For Efficient Processing Of Recurrent Neural Networks 20200429
1734516356601.png


US2021133568A1 METHODS AND SYSTEMS FOR TRAINING MULTI-BIT SPIKING NEURAL NETWORKS FOR EFFICIENT IMPLEMENTATION ON DIGITAL HARDWARE 20191101

Recurrent neural networks are efficiently mapped to hardware computation blocks specifically designed for Legendre Memory Unit (LMU) cells, Projected LSTM cells, and Feed Forward cells. Iterative resource allocation algorithms are used to partition recurrent neural networks and time multiplex them onto a spatial distribution of computation blocks, guided by multivariable optimizations for power, performance, and accuracy. Embodiments of the invention provide systems for low power, high performance deployment of recurrent neural networks for battery sensitive applications such as automatic speech recognition (ASR), keyword spotting (KWS), biomedical signal processing, and other applications that involve processing time-series data.


The research may involve additional NN players other than the usual suspects.
 
  • Like
  • Thinking
  • Fire
Reactions: 11 users

7für7

Top 20
-Hey Mercedes….?

-yes ? How can I help you?

-let AKIDA fix it…

-OK!

4DE6FAE7-98C7-4F75-A554-B0A70F1C95EF.png


Edit: jeeeez ChatGPT…. What a stunning bike!!! Wohoooo
 
  • Fire
  • Like
Reactions: 7 users

Diogenese

Top 20
I’m still excited about what Akida1000 can do. Gen 2 with TENNs must be impressive. How good is this?


APPLICATION​

A single SNAP Card, deployed on an airborne ISR platform, occupying only one 3U VPX slot, could deploy up to five complex machine learning algorithms across each of its onboard processors to perform the following missions simultaneously.

– Automated Target Recognition (ATR) on full motion video feeds and imagery (to include 4k and higher)​

– Real-time detection and identification of threat radars and their acquisition/operating modes​

– Detection of FISINT (Foreign Instrumentation Signature Intelligence) and/or hacking across the airframe’s 1553 communications bus​

– Perform communications analysis to include speech-to-text and foreign language translation of intercepted communications​

– Fuse multiple infrared cameras (e.g., SWIR, MWIR, LWIR) to provide a combined infrared operating picture on the ground Multi-user, simultaneous modulation/demodulation​


View attachment 74421
5 Akidas is a lot of SNN clout!
 
  • Like
  • Fire
  • Thinking
Reactions: 25 users

Boab

I wish I could paint like Vincent
This looks like a pretty good deal. Can the boffins allay my fears. And someone cleverer than me can post the video rather than just the grab.
Screen Shot 2024-12-18 at 7.08.49 pm.png
 
  • Thinking
  • Like
Reactions: 5 users

Rach2512

Regular
Interesting like, sorry if already posted.
 

Attachments

  • Screenshot_20241218-194255_Edge.jpg
    Screenshot_20241218-194255_Edge.jpg
    510.4 KB · Views: 208
  • Screenshot_20241218-194304_Edge.jpg
    Screenshot_20241218-194304_Edge.jpg
    427.4 KB · Views: 206
Last edited:
  • Like
  • Love
  • Fire
Reactions: 35 users

White Horse

Regular
  • Like
  • Love
Reactions: 8 users
Great to see someone (not looked for others yet) in a Dev Community is across Akida. There's a few posts on neuromorphic but haven't flicked through them yet.

Appears he only joined the community early Dec and his first post is below.

DEV Community is a community of 2,548,477 amazing developers​

We're a place where coders share, stay up-to-date and grow their careers.


Rupeshit
Rupeshit
Posted on 13 Dec

Cybersecurity Challenges in AI-Powered Medical Devices (SaMD):​


Excerpt:

Beyond Traditional Boundaries - A New Frontier of Digital Healthcare Protection

The Emerging Threat Landscape

In the converging worlds of artificial intelligence and medical technology, we're witnessing an unprecedented cybersecurity challenge that transcends traditional security paradigms.

AI-powered medical devices are no longer simple data collectors—they've become intelligent, autonomous decision-making systems with potential vulnerabilities that extend far beyond conventional cybersecurity frameworks.

3. Neuromorphic Security Modeling
Develop security architectures inspired by human neural networks:


  • Contextual Anomaly Detection: Design systems that understand not just statistical anomalies, but contextual irregularities in device behavior.
  • Predictive Threat Modeling: Use advanced machine learning to anticipate potential security breaches before they occur.
Neuromorphic Security Modeling: Advanced Approaches

Contextual AI Security Frameworks Tools:


  • IBM's Neuromorphic Computing Research Kit
  • Intel's Loihi Neuromorphic Research Community Edition
  • BrainChip Akida Neuromorphic Processor
Contextual Anomaly Detection Prototype:

There is "code" included in this section.
 
  • Like
  • Fire
  • Love
Reactions: 28 users
This looks like a pretty good deal. Can the boffins allay my fears. And someone cleverer than me can post the video rather than just the grab.
View attachment 74426


I wouldn't worry too much, the guy seems like a bit of a nut job..
Cooks his computer chips in the oven and talks to his dogs?..

Seriously powerful bit of kit there, but totally different competitive environment, I think..

At least that particular unit, isn't going to fit well in a hearing aid or smart ring etc 😛..

Even in robotics, humanoid or otherwise, Neuromorphic is a far better direction, in my humble opinion.



For example, Clone Alpha is following a much more bio-inspired direction, with full skeletal structure, replicated hydraulic muscles (powered by an electric "heart" pump) and nervous system..



When I first saw this, I thought "yeah that's a few years off"

But the first 279 are up for pre-order.


Screenshot_20241219-020242_Firefox.jpg




Everything is moving so quickly, this looks a lot more like Musk's original vision for Optimus and that was people dancing around in leotards..
 
Last edited:
  • Like
  • Wow
  • Fire
Reactions: 13 users

Esq.111

Fascinatingly Intuitive.
This looks like a pretty good deal. Can the boffins allay my fears. And someone cleverer than me can post the video rather than just the grab.
View attachment 74426
Good Morning Boab ,

After a quick search, all I could find was this .

Regards,
Esq.
 

Attachments

  • 20230824_063426.jpg
    20230824_063426.jpg
    5.3 MB · Views: 62
  • Haha
  • Like
  • Fire
Reactions: 14 users
  • Haha
Reactions: 3 users

Esq.111

Fascinatingly Intuitive.
Greetings Chippers,

Strange things are happening in the sky's above the USA .

Just started listening to this interview , if you ask me it SMACKS of our tech.



Regards,
Esq.
 
  • Like
  • Thinking
  • Wow
Reactions: 12 users

Frangipani

Regular
I bet the lucky Master student that gets picked for this position will be working with Akida! 🤗

Airbus Ottobrunn (just outside of Munich) is one of the consortium partners of the Airbus NEURAVIS proposal (the others being Airbus Toulouse, BrainChip, Frontgrade Gaisler & Neurobus) which won the ESA ITT “Evaluation of Neuromorphic Computing Technologies for Very Low Power AI/ML Applications in Space” earlier this year (see my post below):



7E5BB593-3244-42F9-9AC2-1BCCAAFE95C7.jpeg

5F369922-30EC-4745-9AD8-1BE8A1B1673E.jpeg


7151603E-33E7-4C84-BC1F-2B1148470F75.jpeg




https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-429615

CE037482-E5DC-4287-8869-B1466A522F05.jpeg
 
  • Like
  • Fire
  • Love
Reactions: 42 users

7für7

Top 20
Greetings Chippers,

Strange things are happening in the sky's above the USA .

Just started listening to this interview , if you ask me it SMACKS of our tech.



Regards,
Esq.

Noooo… not now… please, no alien invasion right before BrainChip’s breakthrough… 😭 They had so much time… and they’ll still have time in 100 years… 😭 Why now???
 
  • Haha
Reactions: 9 users

Shadow59

Regular
Noooo… not now… please, no alien invasion right before BrainChip’s breakthrough… 😭 They had so much time… and they’ll still have time in 100 years… 😭 Why now???
Don't worry they wont be able to use their technology if its like akida. We are protected by world wide patents!
 
  • Haha
Reactions: 3 users

7für7

Top 20
Don't worry they wont be able to use their technology if its like akida. We are protected by world wide patents!
Ohhhh you’re Right! Thanks for clarifying that! I was worrying for a moment. Also they need a visa first of all to enter the U.S.
 
  • Haha
Reactions: 7 users

buena suerte :-)

BOB Bank of Brainchip
  • Like
Reactions: 7 users

Diogenese

Top 20
Don't worry they wont be able to use their technology if its like akida. We are protected by world wide patents!
But they could get a licence ... think what that would for the share price!
 
  • Fire
  • Haha
Reactions: 8 users

Boab

I wish I could paint like Vincent
Archer Materials (AXE) got a speeding ticket the other day.
Their explanation (see below) has just enhanced the speculation and last I looked it was up another 20% today.
Please Santa more speculative buyers for BRN.

1734570073113.png
 
  • Fire
  • Like
  • Thinking
Reactions: 12 users

7für7

Top 20
Archer Materials (AXE) got a speeding ticket the other day.
Their explanation (see below) has just enhanced the speculation and last I looked it was up another 20% today.
Please Santa more speculative buyers for BRN.

View attachment 74453
Yes! +150% tomorrow with the aliens!!
 
  • Haha
  • Like
Reactions: 3 users
Top Bottom