BRN Discussion Ongoing

[Fullmoonfever]

Couple of things I'm interested in watching.

How the industry starts to react / move given that imo, Prophesee with the wider release of their sensor and hopefully the release of VVDN / BRN edgebox shortly will start to take "neuromorphic" a little more mainstream acceptance than just "research".

And, whether the 2 strike narrative picks up pace at the other place, here and elsewhere by those with possible agendas on the lead up to the AGM.

Personally think a wholesale spill is foolish but each to their own.

I would more prefer that if an individual is not performing that they own it and either there is an adjustment around remuneration etc or they step aside so to speak and accept they and the company maybe aren't the right fit for each other.

If they truly believe they have a pipeline of success coming that they and their team have built, then fine, see it through and get some results.

 

[HopalongPetrovski]

Maybe continuously appear at the companies in person, ring the bell in storm, get on their nerves and repeatedly ask them to accelerate their development and beg to finally buy and pay
I wonder if this is a good concept. I can't follow or agree with some here. Do one just have to be tougher as a salesperson and the other side will work faster?
...
I am completely in your opinion.

Yes.

Desperation is not an endearing quality when you are trying to convert prospects into sales.
Indeed, in order to win business you need to exhibit competence, expertise, competitiveness, longevity and likability.

So much of the process involves developing and maintaining trust and for a new class of product (such as Akida) in a newly developing field, under continuing evolution, we gain much of these qualities from the credentials of our board and senior management.

Once we‘ve had a few generations performing in field our “product” will speak for itself, but until we have achieved that we rely upon the expertise and previously generated networks of our BOD and management to both guide and enlighten our journey with their previous extensive experience.

Neither Rome nor BrainChip will be built in a day.
GLTAH

 

[Guzzi62]

Not sure if posted before.

Prophesee shipping neuromorphic event-based edge vision sensor​

Prophesee shipping neuromorphic event-based edge vision sensor

Peophesee, the French neuromorphic vision sensor specialist, has availability of an event-based vision sensor developed for integration into Edge AI

www.electronicsweekly.com

 

[Diogenese]

Wow!


View attachment 48307

Biden to sign executive order regulating AI use in federal government​

The U.S. president is expected to sign an order on Monday, Oct. 30, to institute guidance for vetting AI tools for government use.

Mike Dalton
Oct. 27, 2023 at 12:30 am UTC
2 min read
Updated: October 27, 2023 at 2:19 am

PHOTO BY GAGE SKIDMORE / CC BY-SA 2.0 / WIKIMEDIA. REMIXED BY CRYPTOSLATE
Share

U.S. President Joe Biden is expected to sign an executive order on artificial intelligence (AI) regulation, the Washington Post said on Oct. 24.
The order, which will likely be signed on Monday, Oct. 30, is said to have at least three goals. First, the order will require advanced AI models to undergo an assessment before they are used by the U.S. government and its federal workers. It is unclear which models this rule might apply to and which criteria the assessment might include.
Second, the order will task select U.S. agencies, including the Department of Defense, Department of Energy, and certain intelligence groups, with examining how they might use AI to improve national cybersecurity.

Biden to sign executive order regulating AI use in federal government

The U.S. president is expected to sign an order on Monday, Oct. 30, to institute guidance for vetting AI tools for government use.

cryptoslate.com

View attachment 48309


View attachment 48308

BrainChip and Quantum Ventura Partner to Develop Cyber Threat Detection

BrainChip and Quantum Ventura partner to develop new cyber threat detection solutions. Learn more about its impact on cyber defense.

brainchip.com

Hi Bravo,

Great find!

The original Phase II SBIR from April 2022 called for a comparison between Akida and Loihi, so the May 2023 announcement ( https://brainchip.com/brainchip-and-quantum-ventura-partner-to-develop-cyber-threat-detection/ ) indicates how that comparison went.

SBIR
https://www.sbir.gov/node/2285619

Realtime Neuromorphic Cyber-Agents (Cyber-NeuroRT)​

Award Information
Agencyepartment of Energy Branch:N/A
ContractE-SC0021562 Agency Tracking Number:0000263950
Amount:$1,650,000.00
Phasehase II
Program:STTR
Solicitation Topic Code:C51-03a
Solicitation Number:N/A
Timeline
Solicitation Year:2021
Award Year:2022
Award Start Date (Proposal Award Date):2022-04-04
Award End Date (Contract End Date):2024-04-03


As part of Phase 1 feasibility study, we evaluated the viability to develop a real-time HPC-scale neuromorphic cyber agent software called Cyber-NeuroRT. We evaluated several scalable neuromorphic techniques to detect and predict cybersecurity threats, compared full precision machine learning models with neuromorphic models and developed an end-to-end Proof of Concept (POC).

Upon completion of Phase 2 prototype, we will produce dramatic reductions in latency and power--up to 100x--without sacrificing accuracy. This will enable quicker response times and savings in operating costs. Cyber-NeuroRT will be a real-time neuromorphic processor-based monitoring tool to predict and alert cybersecurity threats and warnings using the Neuromorphic Platforms of Intel Loihi 1 and BrainChip Akida.

For our Phase 1 POC development, we used 450,000 Zeek log entries with a mixture of normal and malicious data for training the supervised ML models. As part of our study, we covered the following: Cyber Attack types covered – 8 attack types: backdoor, DDOS, DOS, injection, password, ransomware, scanning and XSS, Source files – Zeek log files and Packet Capture Format files (PCAP) containing both malicious and normal records. We used both Supervised and Unsupervised algorithms.

We used algorithms including SNN and CNN-to-SNN conversion with unsupervised learning and supervised learning rules.

To build a full-fledged prototype of Cyber-Neuro RT, we plan to transition the proof-of-concept work to scale to a large data set with additional threat types and other datasets from an HPC environment. HPC environments operate at larger scales than traditional IT domains and our solution should be able to monitor and predict events at more than 160,000 inferences per second.

Tuning of Spike Neural Networks (SNN) parameters such as precision of weights and number of neurons used are two software parameters to explore. The chip can be tuned between high v. low power modes and performance can be studied as a function of power draw.

Evaluation will be performed across a variety of datasets and parameter settings to estimate deployment performance. We will work on efficiency scaling of SNN algorithms in terms of accuracy and hardware metrics like power and energy consumption. Since cybersecurity attack classification is a temporal process, we will leverage recent advancements in the algorithm community to map temporal dynamics of SNNs to recurrent architectures.

Further, to adapt to novel attack vectors, we will explore unsupervised learning techniques in a dynamic network architecture where we will grow or shrink the network as and when novel attack vectors arise. We will also perform an algorithm-hardware co-design analysis by ensuring that our algorithm proposals cater to and consider specific constraints from Akida or Loihi processors like network size, bit quantization levels, among others. 3.1

Some of the features of Cyber-NeuroRT prototype shall include: Ability to monitor, predict and provide system wide alerts of impending cybersecurity threats and warnings at scale by collecting and prioritizing data from Zeek logs and PCAP files streamed in real-time or batch.

We will expand and refine different training techniques like CNN to SNN conversion, direct backpropagation training through surrogate gradient methods or local unsupervised Spike Timing Dependent Plasticity (STDP) enabled approaches.

Compare performance of threat detection between neuromorphic processing vs GPU-based systems and compare between Akida and Intel Loihi processors. Ability to process the data system-wide at an unprecedented scale enabling adaptive, streaming analysis for monitoring and maintaining large-scale scientific computing integrity. Dashboards for security administrators and security analysts.


The project runs to April 2024 which may allow Akida 2 to be taken into account. I guess TeNNs could further improve the speed of response.

 

[DingoBorat]

I wonder what would end the current one sided love affair, between the shorters and BrainChip?

Attributed to Banksy.



You got me runnin' Looks like I'm losing again Feels like I'm losing a friend You got me runnin', ooh You got me runnin'

I'm on a wire Now you're turning to ice Yes, I'm learning the price But I'm on fire, ooh You got me runnin'

Yesterday I bought you flowers And in the night I watched them die Today the flower lady's gone I woke up alone and I Cried. You took me for a ride!..


We all know, that we're the ones who've been taken for a ride lately and that shorters love to take it both ways, but what would it take for them to lose control, of the gift giver that BrainChip has become to them?..

I think the "risk" of that time coming is increasing..

I think they know that..

Their acts of desperation, both on and off the field, will likely increase in the short term.

You have a choice..

 

[IloveLamp]

https://www.embedded.com/how-lidar-is-becoming-smart-and-going-beyond-just-automotive/

 

[Jumpchooks]

https://www.linkedin.com/in/phillipzada/?originalSubdomain=au

‘Facebook for cows’: How a software engineer rustled up an app to tackle cattle thieves

The new technology being trialled at a small property north of Melbourne gives each cow a profile picture electronically linked to its owner.

www.theage.com.au

Facial recognition for cows. This'll fix those Duffas.

Maybe there is a chance that Brainchip would be handy?

 

[Jumpchooks]

https://www.linkedin.com/in/ACoAAA6t0jABZoR83g9Xp7ajNiXG7PIgSFCK044

Brainchips' Akida, Neuromorphic Software.​

Hi Dr Phillip Zada, I have been following the Cattle Duffing problem with interest and see that it needs a Tech Solution, well done! I hope you have heard of Brainchips Akida Neuromorphic Software? It may be ideal for your Stocktake Program, especially as it has the ability to be energy efficient and functions extremely well at the Edge Computing. I am not technical, but I have shares in Brainchip. Please give them a call www.brainchipinc.com

 

[Galaxycar]

I think it is good to air our intentions to vote strike 2, as a shareholder at the next AGM You all would be blindsided and say WTF just happened.

 

[IloveLamp]

Just TATA partnering with APPLE for iphones in INDIA.......ho hum.....nothing to see here

Dr. Sumukh SG on LinkedIn: #tata #apple #iphone #india | 50 comments

Tata Group becomes official manufacturing partner for iPhone in India This is a significant development for India, as it positions the country as a major… | 50 comments on LinkedIn

www.linkedin.com

 

[Fullmoonfever]

TCS mates recently posting about their neuromorphic Small Sat presso at IJCNN not long ago.


Tata Consultancy Services - Research
69,342 followers
1w

Small satellites in space need to transmit data for downstream analysis but are restricted by limited power and on-board computing capacity. At the International Joint Conference on Neural Networks (#IJCNN) 2023, our researchers in collaboration with Indian Institute of Technology, Delhi presented a novel lossless data compression approach based on neuromorphic computing paradigm, that can enable small satellites to compress high resolution Earth observation images before sending them downstream to ground stations for further analysis. Their novel Spiking Neural Network (SNN) based approach, that achieves high compression ratio, is power efficient and has low latency and memory footprint, making it a good fit for the small satellites. Read the #ResearchPaper here- https://bit.ly/45DVrAe Authors- Sayan Kahali; Sounak Dey; Chetan Kadway; Arijit Mukherjee; Arpan Pal; Manan Suri #Research #Publications #Satellites #Computation

Low-Power Lossless Image Compression on Small Satellite Edge using Spiking​

 

[HappyChippy]

Qualcomm eyes local packaging of AI PC chips in India with Tata Group

Qualcomm eyes local packaging of AI PC chips in India with Tata Group

za.investing.com

Made in India Apple iPhone: All you need to know about Tata-Wistron deal

Apple iPhone: The Wistron board reportedly approved the sale of Wistron InfoComm Manufacturing India Private Limited to Tata Electronics.

www.hindustantimes.com

Ineteresting….

 

[Tothemoon24]

Interesting overview of vehicle software , when we launch into a couple of theses it’s game over & we are truly airborne

 

[rgupta]

Interesting overview of vehicle software , when we launch into a couple of theses it’s game over & we are truly airborne

View attachment 48337

If we consider it true that means game is over for arm as well !!!!

 

[manny100]

AI at the Edge via the cloud will not be suitable for any Driving/car motion functions that require real time. Simple non safety related functions AI via the cloud will be ok. Eg open door command, windows down or up eyc. But why would you bother seperating the 2.
From an Investor April'23 BRN Presentation..
Shows major AI players approach to the Edge. ARM, NVIDIA etc have the classical approach via the cloud and is not suitable for EV functions that require real time.
ATM we are in the box seat - for now anyway.

 

[manny100]

Competition analysis as per an April 2023 BRN Investor presentation.
As you can see as at April 2023 BRN was streets ahead at that time.
Looks good.

 

[Bennysmadness]

Do qualcomm still use the Zeroth chip?

 

[toasty]

Thanks for sharing this correspondence.
Some may consider part of Tony’s response mildly price sensitive. Reading between the lines we need to stop talking about the SP for atleast a few months on this forum and only research on where BRN maybe getting revenue before the next AGM in May 24. Interesting times ahead but don’t get alot of confidence from Tony response that BRN management team will avoid a second strike

The thing that worries me about Tony's reply is the use of the word "hope". Hope is NEVER a management strategy and offers me no comfort that the road to revenue is progressing appropriately.

And secondly, I think management is deluding themselves if they think that the 1st strike was about the share price rather than the remuneration report. The big salaries and share bonuses without any transparent communication of tangible results, and not delivering what they said they were going to, are the issues in my opinion at the heart of the vote down.

 

[Schnitzel lover]

Competition analysis as per an April 2023 BRN Investor presentation.
As you can see as at April 2023 BRN was streets ahead at that time.
Looks good.

Tony sells it better than Sean lol.. Thanks for sharing Manny

 

[Bravo]

Do qualcomm still use the Zeroth chip?

Yes, I believe so. For several years Qualcomm and Brain Corp worked together on a program called Zeroth which centred around a neural processing unit (NPU) AI accelerator chip and a software API to interact with the Zeroth platform. It is used for image and sound processing, including speech recognition and the software operates locally rather than on the cloud. The way I understand it is that the cognitive computing abilities developed though the Zeroth program were incorporated into the Snapdragon processor.