BRN Discussion Ongoing

[Cartagena]

Hi Cartegena,

The article is a year old, but it's good to revisit this bit:

"The Vision EQXX’s electronic brain uses something called neuromorphic computing, a form of information processing whose hardware runs so-called ‘spiking neural networks’ that mimic natural ones, firing only when certain thresholds are reached.
Developed in conjunction with California-based artificial intelligence (AI) specialist BrainChip, the system is said to significantly reduce energy consumption. For example, Mercedes claims the “Hey Mercedes” voice control system in the Vision EQXX is five to ten times more efficient than the setup in current production cars.
All this hardware and software makes the Vision EQXX ridiculously easy to drive while achieving extraordinary levels of energy efficiency."

The electronic brain uses SNN. "Hey Mercedes' 5 to 10 times power efficiency improvement is provided as an example of the benefit of Akida in the electronic brain.

It would make sense to use Akida for all in-cabin functions which are amenable to NN implementation - driver monitoring, driver recognition, occupancy, ..., as well as in the ADAS - lidar, ultrasound, camera image classification, ...


... and the bit about the silicon battery is also of interest to us oppressed Talga fans. A lot of premium vehicles will need the 10% Si for the 40% capacity increase.

August 2023
Mercedes-Benz AG

Development engineer in the field of artificial intelligence with a focus on neuromorphic computing and ADAS (f/m/x)​

Tasks​

The Mercedes-Benz Group AG is one of the most successful automotive companies in the world. Together with Mercedes-Benz AG, the vehicle manufacturer is one of the largest providers of premium and luxury cars and vans.
Innovative, forward-looking concepts & products for the future of Mercedes-Benz are being developed in our "Future Software & AI" department in the area of "Innovations & Future Technologies ". The portfolio covers the user experience, software architecture, Metaverse, WEB 3.0 and AI key features of our vehicles and the associated ecosystem. Close collaboration across team boundaries makes it ideally possible for us to realize synergies in the development of an intelligent and inspiring user experience in the vehicle, ecosystem and metaverse. Modern software and AI technologies are seamlessly integrated here
As an employee of the Team AI Research , you are researching current trends in artificial intelligence such as GenAI, quantum AI or neuromorphic computing. We are intensively examining new artificial intelligence methods and working on them further in order to make them usable for our company together with our colleagues from the specialist departments. We strive to expand the limits of artificial intelligence and create customized innovative solutions for our customers. In our team, you can actively shape the future of technology.
These challenges await you:

• 
  
• Applied research in the field of neuromorphic computing and spiking neural networks
• Development of innovative algorithms for automotive applications
• Implementation of neuromorphic algorithms for automotive applications
• Researching and transferring current neuromorphic computing trends for automotive applications with a special focus on ADAS
• Conceptual implementation of ideas and development of demonstrators
• Close cooperation with colleagues from the project and the specialist departments
• Involvement in the development of new ideas and in the development of the strategic orientation of the project
• Close cooperation with universities, research institutions, technology companies and start-ups
• Supervision of internships, bachelor and master thesis and doctoral dissertations

Qualifications​

We need your consent to load the Youtube service!​

We use a third party service to embed video content that may collect data about your activity. Please review the details and accept the service to watch this video.
More InformationAccept
Apply for this job
Print Job-ad

• Job-ad-number:​

  MER0002RHQ

• Publishing date:​

  07.08.2023

• Field of activity:​

  Research and Development incl. Design

• Department​

  Artificial Intelligence Research

To location: Sindelfingen, Mercedes-Benz Plant

Contact​

• Valerij Asmus
  Email: valerij.asmus@mercedes-benz.com

• © 2022 Mercedes-Benz Group AG. Alle Rechte vorbehalten.

 

[Tony Coles]

Brainchip, part 2, TENNs

short video, was expecting more!

 

[Tothemoon24]

Could you soon be running AI tasks right on your smartphone? MediaTek says yes​

MediaTek is working with Meta's Lllama 2 LLM to run generative AI tasks directly on mobile devices, no cloud-based processing required.

For on-device generative AI to access these datasets, mobile carriers would have to rely on low-latency edge networks -- small data centers/equipment closets with fast connections to the 5G towers. These data centers would reside directly on the carrier's network, so LLMs running on smartphones would not need to go through many network "hops" before accessing the parameter data.

Redirecting

 

[Tothemoon24]

The network transformation we have undergone the past few years has brought us to this place of having a highly sophisticated and customizable network able to provide a range of capabilities needed for a variety of devices,” said Adam Koeppe, SVP of Network and Technology Planning at Verizon. “The evolution of this new light software coupled with the technologies deployed throughout our 5G network will allow us to provide the best possible experience for our customers in an efficient manner and will drive the 5G ecosystem with low-cost, energy-efficient devices.”

https://www.ericsson.com/en/press-releases/6/2023/verizon-ericsson-and-mediatek-advance-the-5g-low-cost-low-power-device-ecosystem?utm_medium=social_organic&utm_source=linkedin&utm_campaign=bnew_news_accessnetw_20230824

 

[TECH]

Brainchip, part 2, TENNs

short video, was expecting more!

Gidday Tony,

I viewed this earlier this morning, it's clear to all that the company is ramping things up, as they should be with the general release of AKD 2.0 fast approaching, I'm also expecting some sort of news update on AKD 3.0 over the coming months as I am aware that there's a lot of work going on at US HQ. (like always).

Todays part 2 of the IQP or IQ of Peter (extremely high), the question is, will any new information be shared, as in an late edit over the last week, whatever the case is, Antonio is a must listen, he communicates like Coby Hanoch, extremely well.

 

[Tothemoon24]

https://www.eetimes.com/new-silicon-labs-platform-chases-big-ai-ml-compute-gain/

There was very little detail on the new Series 3 platform itself (you could call this a pre-announcement). But Johnson said things like, “this is a watershed moment,” and “this would increase the adoption of IoT to the next level.” He also emphasized that the new platform wasn’t just an incremental step but a massive step–in terms of enabling 100× improvement in compute performance for AI/ML workloads, more security and greater power efficiency.

In his Q&A session with media, Johnson said that the AI/ML-workload-performance improvement is enabled by architectural design innovation, using proprietary compute engines developed with efficiency for the edge in mind.

 

[Boab]

Brainchip, part 2, TENNs

short video, was expecting more!

2min 8sec "We are generations ahead of the industry"

 

[rgupta]

Brainchip, part 2, TENNs

short video, was expecting more!

Generations ahead of competitors !!

 

[cosors]

Hold on tight.

"In an interview with Yahoo Finance's Jennifer Schonberger on Thursday, Boston Fed President Susan Collins said it is "extremely likely" the central bank will need to hold interest rates high to bring down inflation.

"I think that it's going to take some time to really be sure that we are seeing the sustained realignment of demand and supply that is needed in order to bring inflation back on a path that will get back to 2% [in] a reasonable amount of time," Collins said in an interview from the Jackson Hole Economic Symposium in Wyoming.

"I think it's extremely likely that we will need to hold [interest rates at current levels] for a substantial amount of time," Collins added, "but exactly where the peak is I would not signal right at this point. We may be near [a peak], but we may need to increase a little bit further."

 

[FJ-215]

https://www.eetimes.com/new-silicon-labs-platform-chases-big-ai-ml-compute-gain/

There was very little detail on the new Series 3 platform itself (you could call this a pre-announcement). But Johnson said things like, “this is a watershed moment,” and “this would increase the adoption of IoT to the next level.” He also emphasized that the new platform wasn’t just an incremental step but a massive step–in terms of enabling 100× improvement in compute performance for AI/ML workloads, more security and greater power efficiency.

In his Q&A session with media, Johnson said that the AI/ML-workload-performance improvement is enabled by architectural design innovation, using proprietary compute engines developed with efficiency for the edge in mind.

Very interesting @Tothemoon24.

I liked this bit...


"He added, “Our Series 3 platform is built for a more connected world that demands development flexibility and that more intelligence be pushed to the edge.”

The new platform will move to a 22-nm process technology mode and leverage non-volatile memory tech and off-the-shelf IP blocks."

EDIT:

Edge Impulse is a tech partner of Silicon Labs

Edge Impulse

 

[Learning]

Part 2.



Learning 🏖

 

[robsmark]

Hi Rise

No offense meant, but I'm finding it frustrating reading TSEX lately, you singlehandedly account for what feels like 30% of the posts on this thread over the past 2-3 days.

This isn't a chat room - please take off-topic and low value content, which is almost everything you post, to another thread. I would like to be able to drop in, read a few pages worth of posts in the BRN Discussion thread and feel like I'm up to date on all the happenings of the 1,000 eyes.

You're making it very difficult to do that.

Cheers

Edit. Literally as a write this you've just posted a video for the sake of posting.

If you don’t like it, don’t read it. It’s a public forum, Rise is free to post as much as he wants.

 

[Geared]

Australian patent grant? i cant open the link

 

[Damo4]

(PRICE SENSITIVE ASX ANNOUNCMENT)
New Patent Granted :

BrainChip Granted New Australian Patent Relating to Akida 2.0 IP

 

[supersonic001]

TENNs: A New Approach to Streaming and Sequential Data | BrainChip

TENNs: A New Approach to Streaming and Sequential Data - Check out the new video on TENNs here: https://lnkd.in/gKqAmH2Q - #temperalAI #nueralnetwork #neuromorphic #edgeAI #AI #adas

www.linkedin.com

 

[chapman89]

As I’ve been saying for a long time now that just because there isn’t a license agreement announced does not mean that customers aren’t designing proof of concepts and getting to production with akida, because the customers simply do not want it to be announced due to NDA’s and other confidentiality reasons but people just want sugar hits and not want to wait for the fruit to bear.

Antonio said-
“Every commercial situation is different you know and plus there are confidentiality issues with our partners that we have to be mindful of as well, we can’t forward announce anything that they’re doing, allow me to stress that a little bit more.
When somebody takes a license from us most likely it is way at the front end design cycle and so at best, at best they’re 2 to 3 years out from having production in any sort of meaningful volume, so to be honest they might not even get there so given the time frames plus all the other features that they have to work on that they haven’t done yet at the time they’ve taken a license from us you know our licensee might want to wait a bit before tipping the market with respect to to what technology they’re working on or what product they’ll be producing and naturally this impacts how we make announcements to the market place.
You know for example have you heard of anyone in the world saying that they’re designed into the Galaxy s26 or the iPhone 16? I mean you haven’t heard that it’s too early.
Now I’m pretty confident I don’t think I’m saying anything wrong by saying I’m sure a lot of decisions with respect to those products have already been made but there’s no announcements because they’re not tipping the market yet so there’s a lot of nuances and variances”

 

[Damo4]

View attachment 42855

We should ask Zeebot if we can add this is a reaction!

 

[TECH]

Good morning,

Part 2 of this mornings podcast was once again extremely clear, the questions presented by Tony were right on point, Antonio was deliberate in his answers/comments, as to be expected...great job !

As I have read numerous times, the hardware is the easiest part to get right (always had difficulty understanding that) it's the software that's the time consumer, so having our founder currently over in France overseeing the software team working on MetaTF 2.0 is a blessing, lots of work going on by the entire team worldwide.

His explanation with regards his comment about the share price and what it would/could/will do was beautifully explained, as a number of posters have mentioned, the current valuation is much more in line with where we actually are Cap wise, the foundations that weren't in place to justify the huge market cap we had has and rightfully so, corrected itself, though it hurts me to say that.

"It's not based off expectations and air"....let's all be honest here, we loved seeing a share price north of $1.00/$1.50 and possibly a combination of greed and/or ego we held on tightly to that expectation, when in reality we were all clutching at the empty air in front of our face.

When we do achieve IP Licenses & steady Royalty Income streams it will be all that much more satisfying in knowing that we have arrived and truly are deserving of our place at the big boys table !!

Just keep on topping up, not advice, just a true believers opinion.

God Bless Brainchip....Tech

 

[TheDrooben]

As I’ve been saying for a long time now that just because there isn’t a license agreement announced does not mean that customers aren’t designing proof of concepts and getting to production with akida, because the customers simply do not want it to be announced due to NDA’s and other confidentiality reasons but people just want sugar hits and not want to wait for the fruit to bear.

Antonio said-
“Every commercial situation is different you know and plus there are confidentiality issues with our partners that we have to be mindful of as well, we can’t forward announce anything that they’re doing, allow me to stress that a little bit more.
When somebody takes a license from us most likely it is way at the front end design cycle and so at best, at best they’re 2 to 3 years out from having production in any sort of meaningful volume, so to be honest they might not even get there so given the time frames plus all the other features that they have to work on that they haven’t done yet at the time they’ve taken a license from us you know our licensee might want to wait a bit before tipping the market with respect to to what technology they’re working on or what product they’ll be producing and naturally this impacts how we make announcements to the market place.
You know for example have you heard of anyone in the world saying that they’re designed into the Galaxy s26 or the iPhone 16? I mean you haven’t heard that it’s too early.
Now I’m pretty confident I don’t think I’m saying anything wrong by saying I’m sure a lot of decisions with respect to those products have already been made but there’s no announcements because they’re not tipping the market yet so there’s a lot of nuances and variances”

Agree @chapman89 Antonio could not have explained it any better IMO

Happy as Larry

 

[Bombersfan]

As I’ve been saying for a long time now that just because there isn’t a license agreement announced does not mean that customers aren’t designing proof of concepts and getting to production with akida, because the customers simply do not want it to be announced due to NDA’s and other confidentiality reasons but people just want sugar hits and not want to wait for the fruit to bear.

Antonio said-
“Every commercial situation is different you know and plus there are confidentiality issues with our partners that we have to be mindful of as well, we can’t forward announce anything that they’re doing, allow me to stress that a little bit more.
When somebody takes a license from us most likely it is way at the front end design cycle and so at best, at best they’re 2 to 3 years out from having production in any sort of meaningful volume, so to be honest they might not even get there so given the time frames plus all the other features that they have to work on that they haven’t done yet at the time they’ve taken a license from us you know our licensee might want to wait a bit before tipping the market with respect to to what technology they’re working on or what product they’ll be producing and naturally this impacts how we make announcements to the market place.
You know for example have you heard of anyone in the world saying that they’re designed into the Galaxy s26 or the iPhone 16? I mean you haven’t heard that it’s too early.
Now I’m pretty confident I don’t think I’m saying anything wrong by saying I’m sure a lot of decisions with respect to those products have already been made but there’s no announcements because they’re not tipping the market yet so there’s a lot of nuances and variances”

I agree mate, the next licencees that are announced i would “guess” are much further advanced with products/use cases than what Renesas were when they were announced 3 years ago. Hopefully a much shorter timeframe to market.

I made a comment a while ago that the previous CEO had done a poor job supporting the release and future of the product and Sean has been copping a hammering for it. Between that and the totally wasted 9 months of Peter being CEO he is playing serious catch-up. Antonio basically said exactly that in the podcast.
It sounds like we are really getting there with the overall plan albeit 2 years behind what we thought based on “air” and hype which is mostly what LDN’s business model was.

I thought the podcast part 2 was pretty good.