BrainChip + Tata Consultancy Services

uiux

Regular





BrainChip Holdings Ltd. (ASX: BRN), a leading provider of ultra-low power, high performance edge AI technology, today announced that it has been accepted to present a demonstration featuring its Akida™ Neuromorphic System-on-Chip Technology, recognizing and classifying hand gestures from the audience at the 33rd Conference on Neural Information Processing Systems (NeurIPS), at the Vancouver Convention Center in Vancouver, Canada.

The paper titled Human Gesture Recognition using Spiking Input on Akida Neuromorphic Platform co-authored by Sounak Dey, Arijit Mukherjee from TCS Research at Tata Consultancy Services (TCS) and Gilles Bézard, Douglas McLelland from BrainChip will be presented. The demonstration will involve capturing a few hand gestures and hand positions from the audience using a Dynamic Vision Sensor camera and performing live learning and classification using the Akida neuromorphic platform. This will showcase the fast and lightweight unsupervised live learning capability of the spiking neural network (SNN) and the Akida neuromorphic chip, which takes much less data than a traditional deep neural network (DNN) counterpart, consuming much less power during training.

ss.jpg


---



Application of Spiking Neural Networks for Action Recognition from Radar Data


---


System and method for real-time radar-based action recognition using spiking neural network(snn)

Abstract
This disclosure relates generally to action recognition and more particularly to system and method for real-time radar-based action recognition. The classical machine learning techniques used for learning and inferring human actions from radar images are compute intensive, and require volumes of training data, making them unsuitable for deployment on network edge. The disclosed system utilizes neuromorphic computing and Spiking Neural Networks (SNN) to learn human actions from radar data captured by radar sensor(s). In an embodiment, the disclosed system includes a SNN model having a data pre-processing layer, Convolutional SNN layers and a Classifier layer. The preprocessing layer receives radar data including doppler frequencies reflected from the target and determines a binarized matrix. The CSNN layers extracts features (spatial and temporal) associated with the target's actions based on the binarized matrix. The classifier layer identifies a type of the action performed by the target based on the features.



System and method of gesture recognition using a reservoir based convolutional spiking neural network

Abstract
This disclosure relates to method of identifying a gesture from a plurality of gestures using a reservoir based convolutional spiking neural network. A two-dimensional spike streams is received from neuromorphic event camera as an input. The two-dimensional spike streams associated with at least one gestures from a plurality of gestures is preprocessed to obtain plurality of spike frames. The plurality of spike frames is processed by a multi layered convolutional spiking neural network to learn plurality of spatial features from the at least one gesture. A filter block is deactivated from the plurality of filter blocks corresponds to at least one gesture which are not currently being learnt. A spatio-temporal features is obtained by allowing the spike activations from CSNN layer to flow through the reservoir. The spatial feature is classified by classifier from the CSNN layer and the spatio-temporal features from the reservoir to obtain set of prioritized gestures.

1644578707694.png
 
  • Like
  • Fire
Reactions: 32 users

uiux

Regular
  • Like
Reactions: 4 users

Quiltman

Regular
I believe we should start posting to company based threads within BRN once again.
Great bits of factual evidence & knowledge is lost within the general thread.
So I'll be re-posting information relevant to Tata, TCS and Tata Elxsi since Aug 9th, 2022 ( as I get time ).
Some will be my posts, others copied from other contributors.

And let's try to keep it to facts only - Tata and any of the following - BrainChip, Spiking Neural Networks, Neuromorphic. etc.
You get the idea !

The list below will not be in chronological order - but rather a vault full of treasures.
 
Last edited:
  • Like
  • Love
  • Fire
Reactions: 11 users

Quiltman

Regular
Fact Finder said:
Hi All

As we know Brainchip and Tata Elxsi recently announced their partnership and in doing so issued a press release wherein the following was stated by both companies:

“Tata Elxsi’s partnership with BrainChip will be driving Akida™ technology into medical devices and industrial applications by leveraging BrainChip’s first-to-market, fully digital, neuromorphic technology to provide intelligent, low-power solutions to these demanding target markets.

Akida’s fully customizable, scalable, event-based AI neural processor architecture and its small footprint boosts the efficiency of various applications by orders of magnitude. Greater AI performance at the edge, independent from the cloud, unlocks the growth of the Artificial Intelligence of Things (AIoT) market that is expected to be more than a trillion dollars by 2030.

“The combination of our user-centric design expertise with leading-edge technologies is key to helping enterprises reimagine their products and services to improve operational efficiency, reduce costs and deliver new services to their customers,” said Manoj Raghavan, CEO and MD at Tata Elxsi. “This cannot be possible without our global ecosystem of partners. By partnering with BrainChip and implementing Akida technology into medical and industrial solutions, we are able to deliver innovative solutions at a faster time to market than otherwise possible.””


The above when combined with the following information which I have extracted from the Tata Elxsi website provides me with great comfort as to the exciting prospects for this relationship. Reading the above extract carefully before proceeding to the further information you will take particular note of who stepped forward to comment on the importance of this partnership for Tata Elxsi:

1. Tata Elxsi delivers steady performance in Q1 FY’24 with operating revenue crossing Rs. 850 crores and EBITDA margin at industry leading 29.6%

Continues to build capacity for long-term growth; employee base crosses 12,000 Elxsians

Bangalore, July 17, 2023: Tata Elxsi (BSE: 500408 | NSE: TATAELXSI), amongst the world’s leading providers of design led technology services, announced its first quarter results for the period ending 30th June 2023.

Highlights of the Quarter Ended June 30, 2023:

  • Revenues from operations at Rs. 850.3 Cr
  • Profit before Tax of Rs. 248.3 Cr
  • Crosses 12,000 employee mark with 422 net additions in the quarter
For the quarter ending June 30, 2023, company reported Rs. 850.3 Cr of operating revenue, a growth of 17.1% over the same quarter of previous year (YoY). During the period, Profit before Tax stood at Rs. 248.3 Cr, registering a growth of 8.9% YoY.

Business Highlights:

  • The Industrial Design & Visualization division recorded a strong growth of 41.1% YoY in constant currency terms. Company’s Design-Digital offerings continues to drive volume led growth in a challenging global market environment.
  • Transportation continues to show healthy growth, registering a revenue growth of 17.0% YoY in constant currency terms, and a healthy large deal pipeline across SDV and EV.
  • Healthcare signalled a strong recovery, registering 3.2% QoQ growth in Q1 of FY24 in constant currency.
Mr. Manoj Raghavan, CEO and Managing Director, Tata Elxsi, commenting on the company’s performance in the first quarter of FY24, said:

“We are happy to report a healthy growth of 17.1% YoY in the current macro-economic environment, while delivering industry leading EBITDA margin of 29.6%. While the overall global economic outlook remains challenging, our customer focus and targeted efforts to keep the growth momentum going is showing good results.

During the quarter, our Healthcare & Lifesciences business has reported a healthy QoQ growth of 3.4% which is a significant improvement over the performance during the earlier two quarters. This vertical also reported new product development deal wins for medical diagnostics and Smart Hospital equipment.


https://tataelxsi.com/storage/financial-statements/July2023/TataElxsi-Earnings-Jul17-2023.pdf


2. EXTRACTS FROM SHAREHOLDER PRESENTATION FOLLOWING RELEASE OF 1STQUARTERLY RESULTS 2023 -2024 (FOLLOWED BY QUESTIONS & ANSWERS)

“During the quarter, our Healthcare and Life Sciences business reported a healthy quarter-on quarter growth of 3.2% in constant currency terms, which is a significant improvement over the performance during the earlier 2 quarters. This vertical also reported new product development deal wins for medical diagnostics and smart hospital equipment. In the Transportation business, we continue to see good traction and a strong deal pipeline, especially in software-defined vehicles and electric vehicles.” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED page 2 of 17

“On the people front, we continue to invest in our talent base with a net add of 422 Elxsians in this quarter. The Elxsi family is now 12,000-plus strong” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED page 3 of 17

“Yes. I think I've answered that question. So, the deal pipeline is pretty strong. We still have pretty large opportunities that we have placed bids on and what was supposed to be closed in early part of Q1 is still open. So, customers are definitely taking their time, especially on large bids and so on. So, I think it is natural during the current macroeconomic situation. Customers are being careful before they roll out, especially the large deals and so on. So, I think the fact is that we have a good pipeline, and we see closures happening in Q2. That's the discussion we've had with a number of our customers. we really hope that additional revenues will kick in during Q2. So that's the confidence that we have.” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED Page 5 of 17

“I think over the last 2 quarters, I think the Healthcare and Life Sciences business did not show the growth that we were expecting, and I think that is also because of the transition in the Euro MDR activities and that has taken us a couple of quarters to get over that. But I think this quarter, we have demonstrated good new deals and good product wins there. So that has really helped us to show a good growth. We are pretty optimistic that we will be able to show this continued growth in the subsequent quarters also. But I will not be able to give you a number for the financial year. We don't give that projections. But I think we are on the recovery path, and we hope to grow back to our earlier growth rates in this business pretty soon.” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED Page 6 of 17

Absolutely. This is Nitin here, Bhavik, maybe I'll take that. The way we see it, is twofold. On one hand, in products per se, especially because of the fact that we deal with physical products, whether it's a car, whether it's network equipment, boxes and services on medical equipment, the direct impact of generative AI on such products is very limited. Why? Because they all deal with, if I may use the word specific AI. So therefore, the kind of training that you do, the kind of models that you build, the kind of algorithms that power an autonomous car, or medical diagnosis, or recommendation engine are quite different from what would classically use ChatGPT or generative AI for. So, to that extent, AI becomes more and more relevant. AI becomes more and more predominant within the portfolio of customers, but it's not coming from generative AI per se. In our view, generative AI has some impact on product development to the extent that we are working on how can you aid coding, especially as a starting point rather than starting from scratch, how we can improve testing using generative AI, especially where you had to spend time on creating test cases, describing use case scenarios, there are certain benefits to using generative AI, which are already in the works for us in each of our segments. But the overarching comment would be that in the product engineering business, it has lesser impact. It has far greater impact when you talk of areas like marketing, customer experience, customer care and so on. These are typical BPO/KPO activities. You can straight away imagine that the impact there is far, far higher, and we have no play in those. - MR. NITIN PAI – CHIEF MARKETING AND CHIEF STRATEGY OFFICER Tata Elxsi Limited Page 7 of 17

“Having said that, the medical business is also pretty strong in North America. So, we're really hopeful that, that piece will pick up. So, we look at it, both North America and Europe as very, very key strategic geographies for us. And we continue to invest in those geographies. You would have seen the press releases and we talked about opening a new office in Troy. So, we continue to invest in these geographies. And we are hopeful that growth will return eventually.” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED Page 7 of 17

Sure Satadru. This is Nitin again. So, I'll take that. If you look at the premise on which we, in fact, even built a focus around the medical devices and the health care industry, it was based on the fundamental presumption that in general, the health care industry has been fairly conservative. Product life cycles are very, very long. In fact, there are devices that were designed 15, 20, 25 years back that continue to run. And the advent of cloud, AI connected devices essentially means that you can actually afford to deliver a far more superior customer experience, far more data-led insights that directly affect patient outcomes and health. And therefore, it is essential that all products actually pivot to, in some sense, equivalent of the car. You need to be connected, you need to be automated, I'll not call it autonomous, and you definitely need to look at electronification because a lot of the previous generation medical devices were electromechanical rather than electronic. So that's the premise on which we've built the business, and you'll see that over the last 5 years, from the time we started, we have moved from 0 revenue to about 15%, right? And that's been a business that's grown almost 10% quarter-on-quarter for a few years. So to that extent, I think there is confidence that the industry is actually behind the automotive industry in terms of the transition that it needs to make, even though there are no regulations that are driving it, unlike in the automotive industry, where you have sustainability, you have ESG concerns that put stakes in the ground as far as countries are concerned. This is really an opportunity to change because that is what will sell in the future. That is what will create monetization in the future. So, to that extent, I think there is great confidence that, that pivot has to continue. Software, cloud, IoT, connected devices, intelligence. That is an inexorable path for medical devices. All of them have to get there. The pharma industry has to adopt it to some extent, too, in the sense that as pharma industry moves from prescription-based billing to outcome-based billing, you will need to then monitor patient vitals, you will need to provide for patient tracking and so on. All these are devices that will do it, because the drugs will be then decided based on the markers and that will be personalized based on the actual patient profile. So, in our view, I think the fundamentals of electronics and software becoming the dominant part of what medical devices company you need to do, is a given. The rate that we'll do it at is not driven at the same rate as automotive because automotive has priorities coming from regulatory and regional or country-based decisions that have been made on when they will stop or turn off eyes and so on. The medical device industry does not have that trigger but has a trigger of ‘how do I make money in the future’. So that is what creates absolute confidence. And I think this is just the start of that. We know it's the conservative industry. It's difficult to get in. There are barriers of entry, but we have a marquee set of logos that we work with, we are quite confident of our ability to stay incredibly relevant to these companies.” MR. NITIN PAI – CHIEF MARKETING AND CHIEF STRATEGY OFFICER Tata Elxsi Limited July 17, 2023 Page 9 & 10 of 17

Question: ”Because I'll be very frank, I haven't seen another company like Tata Elxsi doing so diverse verticals and still having so great returns over so many years. So how do you guys look at talent management and these 3 diverse verticals in general?” Answer: So, Nitin, again. Again, just like Coke or KFC, I'm not going to reveal all our secrets. But having said that, maybe I'll start with the fundamental point, right, that when you have been doing this for 25 years and more, and this is the only business that you do, believe me, there's a whole lot of ingrained company knowledge that reflects in its processes and its quality systems in the way we both hire and manage talent. So that's the first fundamental building block that we build on. Equally, we also acknowledge the fact that these 3 industries are so different that the domain knowledge, especially as we move quickly up the seniority and experience levels, demands that you have training programs, you have training methodologies, you have quality systems, you have everything that is different for each of these verticals. And to our advantage, I think that's also a core competency. That's also what makes it very difficult for a generic IT player who depend on fungibility, who depend on training that is industrialized. I mean you would have trainings that people take us for 15 days, 1 month and they are ready for billing. For us, it's nowhere close to that. For us, it's multiples of that timelines to get people ready. But I think that's the way we do business. And hopefully, that's the secret sauce that we bring. “MR. NITIN PAI – CHIEF MARKETING AND CHIEF STRATEGY OFFICER Tata Elxsi Limited Page 11 of 17

“Regarding the Industrial Design business, as I said, it's a smaller business. It's a very, very critical business for us. As compared to last financial year, it has grown significantly. But however, on a quarter-to-quarter basis, we see some amount of flatness and some amount of ups and downs. It's still in the early stages to see how we can have a steady quarter-on-quarter growth there, but that's something that we are working on. We definitely are hopeful that the Industrial Design business will also continue to grow significantly over the next 3 quarters for the financial year.” - Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED Page 12 of 17

“Rajak Kumar Just 2, 3 macro questions. The first one is on the ADAS technology. So, I just want to know, given that we are part of the Tata Group, how do you make insights with other vendors given our exposure with JLR and others, with competition like Samsung who are better placed compared to Tata Elxsi. So how do we overcome this? Nitin Pai: Yes. So maybe I'll take that. Technically, we don't compete with Samsung. We are a potential supplier to Samsung. And when you say Samsung, I presume you mean it in the context of HARMAN. So, to that extent, HARMAN is a customer rather than a competitor. So, in that sense, there is no conflict of interest here. And it's not to do with whether we work with Tata Motors or JLR or not. The question is, do we have capabilities on the software side that add to what they are trying to do in terms of product engineering and product development. And to that extent, we're incredibly relevant. Rajak Kumar: Okay. Got it. And second thing, are we developing any robots linked to the Da Vinci, over on the medical side? Because I've seen your website, it talks about the industrial robotics. So, I just want to know what work we are doing on the medical side? Nitin Pai: Yes. So there is automation - I'm not call it fully robotized surgery or fully automated surgery, but there is a fair amount of work that we're doing, both with AI as well as robotics that has to do with automating and bringing in precision surgery. But at this time, that is all that I can tell you. Rajak Kumar: Currently, that is the only US FDA approved robotic system, right? Nitin Pai: That's right. So, to that extent, that's why I'm hesitant to say that we are working with fully robotized systems. But there is some level of automation and precision surgery that happens in multiple other spaces. Think about eye care, for example. You will see that there are laser guided surgeries that work on cataracts and so on so. It's not that there is no robotic surgery in other areas. If you look at cancer there are equally guided robotic treatments that happen for chemo, radiation and so on. So, I would encourage you to look at is the fact that robotics and automation applies in many levels, just like in cars, from Level 1 to Level 5. Rajak Kumar: Okay. No, my question is more from a strategic standpoint. So, like on a 3-5-year standpoint, will we be challenging Da Vinci systems? Nitin Pai: No, no. So please note, again, like we made the comment on Samsung and so on. We would hope that we work with Da Vinci and create many more Da Vincis by supporting them.” MR. NITIN PAI – CHIEF MARKETING AND CHIEF STRATEGY OFFICER Tata Elxsi Limited July 17, 2023 Page 14 & 15 of 17

3. EXTRACT FROM INVESTOR PRESENTATION SLIDES RELEASED WITH 1ST QUARTERLY REPORT 2023 - 2024

Tata Elxsi won a design-led New Product Development (NPD) deal from a Global Healthcare company to innovate a new line of next-gen Smart Hospital equipment.

********************************************************************************************************************************************


The question becomes how quickly will Tata Elxsi start to deploy Brainchip AKIDA technology solutions in Healthcare or Industrial use cases?

If you believe that the relationship commenced shortly before this recent announcement then the answer is a minimum of one year to a possible a 3 years.

If you believe that Tata Elxsi was introduced to Brainchip by Tata Consulting Services the research arm of the TATA Group of companies then you might believe that the relationship between Brainchip and Tata Elxsi has been in place for a longer period of time and that the announcement has been made because of an impending change of circumstances.

In 2019 Arpan Pal the head of the research group at Tata Consulting Services that has been researching AKD1000 published an interesting paper titled ‘Using Ai on IoT Sensor Data for predicting health of man and machine.’

When it was first published it was publicly accessible but no longer. It is interesting that it was published in December, 2019 at the same time that Brainchip and Tata Consulting Services were demonstrating the use of AKD1000 for identifying hand gestures in real time.

It is also interesting because it covers the two areas Health and Industrial use cases that Tata Elxsi is now partnered with Brainchip to in the words of Manoj Raghavan – MANAGING DIRECTOR AND CHIEF EXECUTIVE OFFICER – TATA ELXSI LIMITED - “implement(ing) Akida technology into medical and industrial solutions, we are able to deliver innovative solutions at a faster time to market than otherwise possible.”

https://www.researchgate.net/public...GFnZSI6InByb2ZpbGUiLCJwYWdlIjoicHJvZmlsZSJ9fQ

The use of the words “faster time to market than otherwise possible” may be highly indicative of a much earlier relationship than the recent announcement implies, even as early as 2020 or 2021. There are clear statements from Brainchip about medical applications being a strong potential growth area for Brainchip dating from at least 2021. In the extracts from the Shareholder presentation I have provided above you will have read about how long Tata Elxsi spends training its people before unleashing them on new applications.

As we well know the AKIDA technology is Science Fiction and no one has yet graduated from the Universities engaged with Brainchip so to make the statement about speed of implementation at the very least Tata Elxsi has finished training its staff. How long would that take well we were told by Douglas Fairbairn that it took MegaChips 18 months to bring their engineers up to speed with the AKIDA technology.

Regards

FF.

Remember do not believe a word I say and DYOR.
 
  • Like
  • Fire
  • Love
Reactions: 11 users

Quiltman

Regular
Building on this ...

Tata, one of the most trusted corporate companies in India, and the 2nd largest, have been working on Neuromorphic computing, and Spiking Neural Networks, for several years.

1695457443222.png



There have been several white papers released by TCS scientists, such as

www.tcs.com

Creating Futuristic Edge Systems with Neuromorphic Computing

As AI techniques on classical processor architecture become a concern, explore neuromorphic computing to create futuristic systems with intelligence at the edge.
www.tcs.com

Some key scientists at TCS even have in built into their work title !

Sounak Dey
Senior Scientist, Neuromorphic Computing and Spiking Neural Network, TCS Research

And this forum has direct evidence of Akida being used in the development of various projects within TCS

Such as the ECG classifier for Wearable edge devices, published 10 months ago.

;
1695457918760.png


We must also remember this LinkedIn commentary from 10 months ago

1695458127612.png


1695458180798.png



In response to a post by TCS on Neuromorphic Computing coupled with Spiking Neural Networks, our own Jesse Chapman got involved and made a comment about BrainChip to Arijit Mukherjee, TCS principal scientist.

Arijit's response :

Jesse : " Brainchip very far ahead of all companies working on neuromorphic, be interesting to see the adoption of neuromorphic architecture through their partners"

Arijit : " We have worked very closely and will continue to do so "

This is not dot joining , 10 months ago it was stated by TCS that they had been working very closely with BrainChip .
and recently they have taken this to the next level via the Tata Elxsi partnership.

The second biggest company in India is right behind creating commercial value via neuromorphic compute coupled with spiking neural networks using BrainChip as a partner.
 
  • Like
  • Fire
  • Love
Reactions: 16 users

Quiltman

Regular
How about this from TATA Senior Scientist on Neuromorphic compute ... given one of the 2 declared "partnership " directions was medical devices.

Tata Elxsi’s partnership with BrainChip will be driving Akida™ technology into medical devices and industrial applications by leveraging BrainChip’s first-to-market, fully digital, neuromorphic technology to provide intelligent, low-power solutions to these demanding target markets.

Published 9 months ago :

1693285499081.png


Chart in detail

https://neurips.cc/media/PosterPDFs/NeurIPS 2022/60044.png?t=1668067062.275098
Bottom Right :

1693285540219.png


Guessing benchmark on BrainChip Akida went pretty damn well given todays announcement !

1693285321138.png


And this white paper from TATA from 1 year ago, which had Sounak as one of the authors.

1693286454863.png


They were all over SNN advantages for a while ...

www.tcs.com

Creating Futuristic Edge Systems with Neuromorphic Computing

As AI techniques on classical processor architecture become a concern, explore neuromorphic computing to create futuristic systems with intelligence at the edge.
www.tcs.com
 
  • Like
  • Fire
  • Love
Reactions: 9 users

Quiltman

Regular
It was almost 4 years ago we announced this joint demonstration with Tata, which in turn would mean BrainChip was working with TCS sometime before this announcement.

In other words, Tata & BrainChip have been working together for at least 5 years.

I suspect now that a commercial arrangement has been announced (ie. " partnership " ), commercial products will materialise relatively quickly, as the groundwork has already been done.

1695542031224.png
 
  • Like
  • Fire
Reactions: 11 users

Quiltman

Regular
Here is the link to the paper titled

“Advancing edge computing capabilities with neuromorphic platforms”​

See link here to the paper- https://www.tcs.com/what-we-do/rese...omorphic-edge-computing-embedded-intelligence

This is the paper by TATA written by these 3 below-

ARIJIT MUKHERJEE​

Principal Scientist, TCS Research​

SOUNAK DEY​

Senior Scientist, TCS Research​

VEDVYAS KRISHNAMOORTHY​

Business Development Manager, TCS​

72F4E482-8876-4DED-A50A-2028311E1EA2.jpeg


The opinions and research I share are my own and I am not licensed. External links are not recommended. To be safe, conduct your own research or seek advice from a licensed financial advisor.

Stock Disclosure:
Hold
Like Reply
Report
  • Like
  • Love
  • Fire
Reactions:35 users

[IMG alt="Terroni2105"]https://thestockexchange.com.au/data/avatars/m/0/11.jpg?1643850676[/IMG]

Terroni2105

Founding Member​

Great find and post @thelittleshort Do you know what was the original question he was answering?
Regards
FF

AKIDA BALLISTA
1669716213789.jpeg


The opinions and research I share are my own and I am not licensed. External links are not recommended. To be safe, conduct your own research or seek advice from a licensed financial advisor.

Stock Disclosure:
Held
Like Reply
Report
  • Like
  • Fire
  • Love
Reactions:38 users

[IMG alt="chapman89"]https://thestockexchange.com.au/data/avatars/m/0/27.jpg?1644040687[/IMG]

chapman89

Founding Member​

Thanks for sharing @Terroni2105.

We’ve known for a while Brainchip and TATA have been working together, through the hand gesture video which Brainchip & TATA collaborated together, and also from previous interactions I’ve had with TATA earlier in the year which I shared here.
But if anybody needs more convincing that TATA is a EAP, then here it is below, this was just 10 minutes ago-

456DC880-C24A-4BB5-B0EA-BE834F714054.jpeg
 
  • Like
Reactions: 9 users

Quiltman

Regular
1696388600462.png
 
  • Like
  • Fire
Reactions: 7 users

Quiltman

Regular

Brainchip partners with Tata Elxsi to drive new opportunities​

Laguna Hills, Ca (USA)​

Brainchip Holdings Ltd, the world’s first commercial producer of ultra-low power, fully digital, event-based, neuromorphic AI IP, has partnered with leading design and technology services provider Tata Elxsi, as an addition to the company's Essential AI ecosystem.

With more than 10,000 employees and multiple locations worldwide, Tata Elxsi is a global leader in providing design and technology services across a multitude of industries and markets. The company helps some of the largest corporations in the world reimagine their products and services through design thinking and the application of digital technologies through next-generation AI solutions.

Tata Elxsi’s partnership with Brainchip will be driving Akida technology into medical devices and industrial applications by leveraging Brainchip’s first-to-market, fully digital, neuromorphic technology to provide intelligent, low-power solutions to these demanding target markets.
Akida’s fully customisable, scalable, event-based AI neural processor architecture and its small footprint boosts the efficiency of various applications by orders of magnitude. Greater AI performance at the edge, independent from the cloud, unlocks the growth of the Artificial Intelligence of Things (AIoT) market that is expected to be more than a trillion dollars by 2030.

“The combination of our user-centric design expertise with leading-edge technologies is key to helping enterprises reimagine their products and services to improve operational efficiency, reduce costs and deliver new services to their customers,” said Manoj Raghavan, CEO and MD at Tata Elxsi. “This cannot be possible without our global ecosystem of partners. By partnering with Brainchip and implementing Akida technology into medical and industrial solutions, we are able to deliver innovative solutions at a faster time to market than otherwise possible.”

“Brainchip is very aligned with Tata Elxsi’s mission to innovate with leading edge technology and deliver compelling new products and services that improve customer experience and outcomes,” said Rob Telson, Vice President of Ecosystems & Partnerships at Brainchip. “Our partnership with Tata Elxsi leverages Akida technology to transform applications and results in markets such as healthcare and industrial automation. We look forward to working with them to create new opportunities and drive growth.”
 
  • Like
  • Fire
Reactions: 7 users

Quiltman

Regular
I see our mates at TCS been doing some work on CFDs and SNN with a presso at Dec 22 NeurIPS.

Looks like they intend to run their algo on Akida in future work.

Paper HERE

Stabilization and Acceleration of CFD Simulation by Controlling Relaxation Factor Based on Residues: An SNN Based Approach

Mithilesh Maurya, Dighanchal Banerjee, Sounak Dey, Dilshad Ahmad
TCS Research, India

Abstract
Computational Fluid Dynamics (CFD) simulation involves the solution of a sparse system of linear equations. Faster convergence to a physically meaningful CFD simulation result of steady-state physics depends largely on the choice of optimum value of the under-relaxation factor (URF) and continuous manual monitoring of simulation residues. In this paper, we present an algorithm for classifying simulation convergence (or divergence) based on the residues using a spiking neural network (SNN) and a control logic. This algorithm maintains optimum URF throughout the simulation process and ensure accelerated convergence of the simulation. The algorithm is also able to stabilize and bring back a diverging simulation to the converging range automatically without manual intervention. To the best of our knowledge, SNN is used for the first time to solve such complex classification problem and it achieves an accuracy of 92.4% to detect the divergent cases. When tested on two steady-state incompressible CFD problems, our solution is able to stabilize every diverging simulation and accelerate the simulation time by at least 10% compared to a constant value of URF.

Conclusion and Future Works. Our proposed solution is generic and can be integrated with any CFD tools. This automated process saves a lot of manual effort and time thereby making it beneficial for many industry scale CFD problems that run for days. Moreover, the SNN model can easily be retrained with additional data to improve the accuracy and to cater to different classes of problems.

In future, we intend to improve the existing fixed control logic to an adaptive URF controller varying with the SR value. We also intend to test the performance and power consumption of the SNN by
running it on a real neuromoprhic hardware such as
Intel Loihi [2, 9] or Brainchip Akida [1].
 
  • Like
  • Fire
Reactions: 8 users

Quiltman

Regular
More on the most recent TCS research paper. I like it a lot!

ieeexplore.ieee.org

Low Power & Low Latency Cloud Cover Detection in Small Satellites Using On-board Neuromorphic Processors

Emergence of small satellites for earth observation missions has opened up new horizons for space research but at the same time posed newer challenges of limited power and compute resource arising out of the size & weight constraints imposed by these satellites. The currently evolving...
ieeexplore.ieee.org
ieeexplore.ieee.org

Bringing Touch to the Edge: A Neuromorphic Processing Approach For Event-Based Tactile Systems

The rise of neuromorphic applications has highlighted the remarkable potential of biologically-inspired systems. Despite significant advancements in audio and visual technologies, research directed towards tactile sensing has not been as extensive. We propose a neuromorphic tactile system for...
ieeexplore.ieee.org

Attachments​

 
  • Like
  • Fire
Reactions: 8 users

Quiltman

Regular
This is an industry I hadn’t considered before but TCS believe AI ML will be part of its future.

www.tcs.com

Embracing Neural Traits for Paint & Coating Manufacturing

Read how next-gen P&C manufacturers must develop neural traits & leverage latest technologies like data & analytics for agility & intelligent decision making.
www.tcs.com
We believe paint and coatings companies must focus on making their manufacturing operations flexible, agile, and resilient, especially after the pandemic, as reducing costs and improving productivity have become crucial to avoid squeezing margins. Smart factories equipped with a well-connected ecosystem of people, process, and technology impart a cognitive capability, where self-monitoring and controlling critical parameters by equipment make autonomous operations a reality. Paint and coatings manufacturers can also leverage AI and ML to receive intelligent and timely alerts on diagnostic, predictive, and prescriptive maintenance for improving machine availability, quality, and throughput. A good example is that of Asian Paints: they created a digital twin for production processes using AI on the data collected from various assets. Digitalization reduced the cycle time for the firm by 7%.

I couldn’t find a date for the above article but it was attached to an article by TATA on the future of Spiking Neural Networks - the building blocks of AI innovation 😀

www.tcs.com

Spiking neural networks


www.tcs.com

The opinions and research I share are my own and I am not licensed. External links are not recommended. To be safe, conduct your own research or seek advice from a licensed financial advisor
 
  • Like
  • Fire
Reactions: 5 users

Quiltman

Regular
Evening team, can't remember seeing this.

TCS using Akida for spacetech, 'cloud cover detection' and prob 'lossless image compression' on small satellites.

View attachment 35514
View attachment 35513
Two great indicators:

1. Gilles Bezard (BrainChip) is an author for the cloud cover satellite;
2.1 Arindam Basu, professor, City Uni HK, asks about BrainChip for the SNN in the edge satellite;
2.2 Airijit Mukherjee replied "Yes. The cloud cover detection work is entirely on Akida.".

It would be nice to see the BrainChip blog/Linkedin ... give this achievement a bit of publicity when the papers are published.
 
  • Like
  • Fire
Reactions: 9 users

Quiltman

Regular
I noticed Tata was brought up today and stumbled across this which might be interesting

Tutorials – 2023 IEEE International Conference on Acoustics, Speech and Signal Processing


2023.ieeeicassp.org
2023.ieeeicassp.org

Tutorials – 2023 IEEE International Conference on Acoustics, Speech and Signal Processing


2023.ieeeicassp.org
2023.ieeeicassp.org

Presenters

Manan Suri (IIT Delhi), Sounak Dey (Tata Consultancy Services Ltd.), Arun M. George (TCS Research & Innovation)


  1. Application & Implementation:
Third and final part of the tutorial will provide a detailed landscape of the applications that have been developed and tested so far using SNN and NC, including our own experiences. Aspects such as spike encoding techniques, conversion of ANNs to SNNs, FPGAs will be discussed.
 
  • Like
  • Fire
Reactions: 3 users

Quiltman

Regular
B14211D2-B0B4-4E63-BF1F-67604F0214D9.png


INTRODUCTION

Embedding intelligence at the edge has become a critical requirement for many industry domains, especially disaster management, healthcare, manufacturing, retail, surveillance, remote sensing etc. Classical Machine learning or Deep learning (ML/DL) based systems, being heavy in terms of required computation and power consumption, are not suitable for Edge devices such as robot, drones, automated cars, satellites, routers, wearables etc. which are mostly battery driven and have very limited compute resource. Inspired from the extreme power efficiency of mammalian brains, an alternative computing paradigm of Spiking Neural Networks (SNN) also known as Neuromorphic Computing (NC), has evolved with a promise to bring in significant power efficiency compared to existing edge-AI solutions. NC follows non-von Neumann architecture where data and memory are collocated like brain neurons and SNNs handle only sparse event-based data (spikes) in asynchronous fashion. Inherently SNNs are very efficient to understand features in temporally varying signals and is found to efficiently classify/process auditory data, gestures/actions from video streams, spot keywords from audio streams, classify & predict time series from different sensors used in IoT, regenerate temporal patterns etc. The community is pursuing multiple sophisticated dedicated Neuromorphic hardware platforms such as: Intel Loihi, IBM TrueNorth, Brainchip Akida, SpiNNaker, DYNAPs to name a few.

Moreover, ultra-advanced and futuristic nanoelectronic devices and materials are being explored to build energy efficient neuromorphic computers. So this domain, as well as this tutorial, lines in the intersection of Computational Neuroscience, Machine Learning and In-memory Neuromorphic Computation techniques.



RATIONALE AND STRUCTURE OF THE TUTORIAL

The ICASSP community is at the forefront of research in the domain of signal processing. Thus it is extremely relevant to conduct a tutorial on advances in the domain of Neuromorphic Computing at the forum. We are proposing two valuable cross vertical elements in this tutorial:



(i) Firstly, the proposed tutorial has been developed keeping both academic and industrial/application interests in mind. The speakers represent leading academic and industry research teams on the subject with several years of theoretical and applied experience on the topic. Over last few years, while solving customer requirements related to edge computing, TCS Research has successfully taken neuromorphic research to real market applications. At the same time, group at IIT-Delhi has contributed significantly towards development of cutting-edge neuromorphic hardware and memory-inspired computing.

(ii) Secondly, the proposed tutorial will not only cover foundational basics (i.e. algorithms, bio-inspiration, mathematics) of the subject, but will also delve in to real hardware-level implementation and actual application use-cases (such as gesture recognition in robotics, time series classification and prediction in IoT, continuous health monitoring, remote sensing via satellite etc.) as pursued in industry so far.



The tutorial is structured to cover all relevant aspects of SNN and NC as detailed below.



  1. Biological Background, Software & Simulation of SNNs:
We will start the tutorial by giving a background of modus operandi of biological neurons, their equivalent computational models, spike generation process, synaptic weight updates and learning rules. Next, we will speak in detail about (i) existing software tools and SNN simulators, (ii) how to create a basic SNN, feed data into it and do a simple classification task, (iii) how to tune it towards better performance.

Speaker: Sounak Dey, duration: 50 minutes.



  1. Neuromorphic Hardware Basics:
Second part of the tutorial will cover the basics of dedicated hardware approaches for implementing neuromorphic algorithms in real world. Key techniques will be bench-marked for their pros and cons. How specialized neuromorphic hardware can offer performance benefits will be discussed. The session will end with discussion of futuristic nanomaterials proposed for neuromorphic computation.

Speaker: Manan Suri, duration: 50 minutes.



  1. Application & Implementation:
Third and final part of the tutorial will provide a detailed landscape of the applications that have been developed and tested so far using SNN and NC, including our own experiences. Aspects such as spike encoding techniques, conversion of ANNs to SNNs, FPGAs will be discussed.

Speaker: Arun George, duration: 50 minutes.



A flexible and interactive model of discussion and Q-A with the audience will be followed throughout the tutorial. Duration:10-30 minutes.
 
  • Like
  • Fire
Reactions: 12 users

Quiltman

Regular

Tata Technologies Achieves 18% Reduction in Power Consumption Using AI/ML - Techiexpert.com

In a pioneering initiative, Tata Technologies has harnessed the capabilities of artificial intelligence and machine learning to revolutionize its
www.techiexpert.com
www.techiexpert.com

The opinions and research I share are my own and I am not licensed. External links are not recommended. To be safe, conduct your own research or seek advice from a licensed financial advisor.

Stock Disclosure:
 
  • Like
  • Fire
  • Love
Reactions: 16 users
Top Bottom