BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Fullmoonfever
Top 20
Ahh....hindsight is such a wonderful thing to give clues as to how long some things could have been brewing.
From a couple of months ago.
www.nextplatform.com
Software, Not Hardware, Will Drive Quantum and Neuromorphic Computing
Jeffrey Burt
2 months ago
As with the Quantum SDK, Intel initially had no intention of creating a software framework out of the software developed and used in-house for neuromorphic computing, according to Mike Davies, senior principal engineer and director of the chip maker’s Neuromorphic Computing Lab. However, it became clear that the lack of a general-purpose software stack was hobbling the industry’s efforts in the field.
“Until Lava, it’s been very difficult for groups to build on other groups’ results even within our own community because software tends to be very siloed, very laborious to construct these compelling examples,” Davies told journalists. “But as long as those examples are developed in a way that cannot be readily transferred between groups and you can’t design those at a high level of abstraction, it becomes very difficult to move this into the commercial realm where we need to reach a broad community of mainstream developers that haven’t spent years doing PhDs in computational neuroscience and neuromorphic engineering.”
Lava is an open-source framework with permissive licensing, so the expectation is that other neuromorphic chip manufacturers – which include the likes of IBM, Qualcomm, and BrainChip – will port Lava to their own frameworks. It’s not proprietary, though Intel is the major contributor to it, Davies said.
The latest iteration of Lava is version 0.5, though the company has been steadily rolling out releases of Lava to GitHub, he said. It also illustrates, along with circuit improvements in Loihi 2, the advancements made in the chip for running deep feed forward neural networks, a basic type of network used in some supervised learning uses. There are fewer chip resources needed to support these networks, the inference operation is up to 12X faster than Loihi 1, and 50X more energy efficient.
These aren’t the types of workloads that Loihi was designed to support – GPUs and other accelerators can run deep feed forward networks well. And Intel is aware of this. “It is very important that we support feedforward neural networks of the kind that everyone is using and loves today because they’re just an important building block for future neuromorphic applications.” Davies said.
From a couple of months ago.
Software, Not Hardware, Will Drive Quantum and Neuromorphic Computing
In emerging computing fields like quantum computing and neuromorphic computing, hardware usually grabs the lion’s share of attention. You can see the
www.nextplatform.com
Software, Not Hardware, Will Drive Quantum and Neuromorphic Computing
Jeffrey Burt
2 months ago
As with the Quantum SDK, Intel initially had no intention of creating a software framework out of the software developed and used in-house for neuromorphic computing, according to Mike Davies, senior principal engineer and director of the chip maker’s Neuromorphic Computing Lab. However, it became clear that the lack of a general-purpose software stack was hobbling the industry’s efforts in the field.
“Until Lava, it’s been very difficult for groups to build on other groups’ results even within our own community because software tends to be very siloed, very laborious to construct these compelling examples,” Davies told journalists. “But as long as those examples are developed in a way that cannot be readily transferred between groups and you can’t design those at a high level of abstraction, it becomes very difficult to move this into the commercial realm where we need to reach a broad community of mainstream developers that haven’t spent years doing PhDs in computational neuroscience and neuromorphic engineering.”
Lava is an open-source framework with permissive licensing, so the expectation is that other neuromorphic chip manufacturers – which include the likes of IBM, Qualcomm, and BrainChip – will port Lava to their own frameworks. It’s not proprietary, though Intel is the major contributor to it, Davies said.
The latest iteration of Lava is version 0.5, though the company has been steadily rolling out releases of Lava to GitHub, he said. It also illustrates, along with circuit improvements in Loihi 2, the advancements made in the chip for running deep feed forward neural networks, a basic type of network used in some supervised learning uses. There are fewer chip resources needed to support these networks, the inference operation is up to 12X faster than Loihi 1, and 50X more energy efficient.
These aren’t the types of workloads that Loihi was designed to support – GPUs and other accelerators can run deep feed forward networks well. And Intel is aware of this. “It is very important that we support feedforward neural networks of the kind that everyone is using and loves today because they’re just an important building block for future neuromorphic applications.” Davies said.
Fullmoonfever
Top 20
Funny how we were discussing the cybersecurity topic and Cyber-Neuro RT earlier today.
Just saw this advertised from a couple of weeks ago.
Scroll to the end. Still in the cyber security game it appears
neurotechx.com
Quantum Ventura
Website
Headquartered in the heart of Silicon Valley’s historic San Jose downtown, Quantum Ventura is in the business of creating innovative and groundbreaking components, systems, and technologies, with the mission of delivering advanced customer- centric solutions. Quantum Ventura’s R&D division, “QuantumX Research Labs”, provides cutting edge technology solutions to federal government agencies and corporations throughout the U.S. We excel in developing concepts into market-focused products and customer-driven solutions.
Just saw this advertised from a couple of weeks ago.
Scroll to the end. Still in the cyber security game it appears
Internship Position in AI, ML and MetaOptics - Part time/ Full time - Quantum Ventura
Tremendous research opportunity(ies) is (are) available to work on cutting edge AI/ML AND/OR metamaterial optics (meta- optics) projects for self-motivated part time or full time candidates open to learning new technologies. For a three to six month engagement [extendable], at 20 to 40 hours per...
neurotechx.com
Internship Position In AI, ML And MetaOptics – Part Time/ Full Time – Possible Remote (San Jose, CA)
- FULL TIME
- Internship
- Part Time
- San Jose, CA, USA
- Posted on November 23, 2022
Quantum Ventura
Website
Internship Position in AI, ML and MetaOptics
San Jose, CA · Part time/ Full timeQuantum Ventura
Quantum Ventura Inc pursues applications of sophisticated research to problems in Artificial Intelligence / Machine Learning, AI / ML Verification & Validation, Image Processing and Classification with Metamaterial Optics, Computational Electromagnetics, Synthetic Aperture Radar (SAR), Cybersecurity, Secure Mobile technology (QuantumX / Diamond-Droid), and HPC-driven Big Data Analytics. Some of Quantum’s clients include the Department of Energy, DARPA, the Department of Defense, the Navy, and the Defense Health Agency. We design creative solutions and build unique products for challenging problems, with complete end-to-end solutions and with unsurpassed technical expertise.Headquartered in the heart of Silicon Valley’s historic San Jose downtown, Quantum Ventura is in the business of creating innovative and groundbreaking components, systems, and technologies, with the mission of delivering advanced customer- centric solutions. Quantum Ventura’s R&D division, “QuantumX Research Labs”, provides cutting edge technology solutions to federal government agencies and corporations throughout the U.S. We excel in developing concepts into market-focused products and customer-driven solutions.
Internship(s) Description
Tremendous research opportunity(ies) is (are) available to work on cutting edge AI/ML AND/OR metamaterial optics (meta- optics) projects for self-motivated part time or full time candidates open to learning new technologies. For a three to six month engagement [extendable], at 20 to 40 hours per week, the candidate will work on projects associated with designing, developing, prototyping, and testing advanced research topics in deep learning AND/OR simulation, optimization, inverse design, and testing of metasurface photonic / meta-optic devices. The work can potentially lead to filing patent applications and publishing technical papers.
Qualifications
Part time or full time candidates or students pursuing or who have completed a Master’s degree or Doctorate in Computer Science, Electrical Engineering, Material Science, Applied Physics, Physics or other relevant disciplines with a background in deep learning AND/OR knowledge of electromagnetism, optics, image processing / computer vision, and the relevant math (e.g., linear algebra, partial differential equations, statistics) and numerical simulations, is desired. Additional requirements are as follows:Machine learning/deep learning skills
- Strong programming and math/physics skills;
- Experience in quick prototyping and scientific research methods (e.g., reading technical papers);
- Experience in Python, pandas, numpy, scikit learn, etc.;
- Experience in deep learning frameworks including Tensorflow, PyTorch, or Keras;
- Experience in machine learning including data cleaning, visualization, training from scratch, fine-tuning classifiers, etc.;
- Experience in neuromorphic computing (e.g., Intel Loihi or BrainChip Akida) or cybersecurity is a plus.
SERA2g
Founding Member
Wait what. I played a few cs 1.6 lan comps too 
next perth catch up let’s definitely discuss!
next perth catch up let’s definitely discuss!
Sirod69
bavarian girl ;-)
LinkedIn Login, Sign in | LinkedIn
Login to LinkedIn to keep in touch with people you know, share ideas, and build your career.
Take a look at what Mohamed Awad had to say about the evolution and trajectory of devices built #onArm in conversation with Jon Masters.
Discover more about the future of Arm in Mark Hambleton’s #ArmDevSummit keynote:
Virtual Hardware – Software Development Without Hardware
Arm Virtual Hardware delivers models of Arm based SoCs to enable software development and testing ahead of hardware availability, accelerating time-to-market for IoT and ML applications.
www.arm.com
cosors
👀
I'll refer the comment to myself. What does machine learning have to do with Akida and BRN
A thought that made me post this little story. Why the Chinese don't let their tool learn about neuromorphic computing is beyond me. But as far as the whole topic is concerned, they have huge problems right now anyway. It's negligible if a Chinese chat bot that doesn't learn enough by itself doesn't know anything about BRN.FlipDollar
Never dog the boys
I wrote that after a bottle of 2016 Craggy Range Sophia.
Please ignore the post, but don’t ignore a bottle of Sophia if you ever get the opportunity.
Have a good day
Sirod69
bavarian girl ;-)
Rochester Institute of Technology- Computer engineering becomes part of inaugural program focused on neuromorphic technologies - powered by BrainChip -
www.rit.edu
Computer engineering becomes part of inaugural program focused on neuromorphic technologies
RIT recently became one of the inaugural academic partners in the BrainChip University AI Accelerator Program. As part of the partnership, RIT’s computer engineering program will receive hardware as well as lecture modules for classes detailing how the novel chips can be programmed and used to...
Sirod69
bavarian girl ;-)
BrainChip will help advance innovation on Intel’s foundry manufacturing platform. The fully customizable event-based AI neural processor enables inference and learning at the edge. Its scalable architecture and small footprint boosts efficiency by orders of magnitude — supporting up to 1024 nodes that connect over a mesh network. https://lnkd.in/djXzkw44 via eeNews Europe
www.eenewsembedded.com
BrainChip joins Intel Foundry Services to drive Neuromorphic AI
Bringing Neuromorphic AI at the edge to IFS, BrainChip will help advance innovation on Intel’s foundry manufacturing platform.
www.eenewsembedded.com
supersonic001
Regular
BrainChip on LinkedIn: BrainChip joins Intel Foundry Services to drive Neuromorphic AI
BrainChip will help advance innovation on Intel’s foundry manufacturing platform. The fully customizable event-based AI neural processor enables inference and…
FrederikSchack
Regular
GrAI Matter competition?
Like others, I came across some impressive numbers from GrAI Matter, so decided to look into what they are and if they are serious competitors.I can see that it was earlier mentioned that GrAI Matter doesn´t sell IP, which is probably not correct, at least not anymore:
Brainchip has four IP Vendors:
- Rensas
- Megachips
- ARM
- Intel
- ARM
- Synopsys
Key differences
Diogenese have explained some of the key differences between GrAI Matter and Akida from Brainchiphttps://thestockexchange.com.au/threads/brn-discussion-2022.1/post-20384
https://semiengineering.com/spiking-neural-networks-research-projects-or-commercial-products/
It seems like Akida is ahead in regards to working independently from the Internet, learning on the edge and can also do recognition with high probability with just 4 bits, where GrAI matter have decided on 8 or 16 bits, which is costing a lot more compute resources and apparently, they are more limited with regards to input and needs to compare pixels from picture to picture. On the other hand, the GrAI VIP has 18 million neurons and it might not matter so much in regards to power consumption that the die is bigger, when it´s a spiking neural network and it´s not using energy while inactive.
Overall I think we have too few details and there are to big differences in how they are working, to really say which chip is the better neuromorphic processor. Most likely they are going to have each their strengths and might both have niches that they’ll dominate and have some areas they overlap. From the few details I have seen, it seems like Akida is much better suited for the edge and that´s the most interesting market to break through in.
I read somewhere that GrAI Matter Labs initially didn´t focus on the edge, but since have changed their focus.
Popularity
To get a better idea of the popularity of the two chips, I did some searches and it seems like Brainchip is factors more exposed on the web, which can of course partly be explained by a very big and active investor group, which GrAI Matter can´t have as they are a private company.“GrAI Matter” returns 94 results in Google search engine
“Brainchip” returns 286.000 results in Google search engine, but of course some also write about "Brainchip" regarding Neuralink which is a misconception, but I think the result is clear anyway.
Google Trends show that GrAI Matter was trending back in Googles infancy, but now it appears that nobody really searches for it, while Brainchip is much more popular.
I also looked at domain traffic with Semrush and I don´t know how reliable those data are, but it seems like there are some organic traffic for GrAI Matter. Brainchip has almost 22 times more organic traffic, but this could be explained by a large and active investors group.
After looking into it, my feeling is that GrAI matter does have something going for it, they are integrated with two IP vendors, but they probably have to do a lot more marketing to really break through and they will probably find it hard to compete with Akida on the edge.
alwaysgreen
Top 20
Excellent work looking into this. Thank you
I wouldn't give too much weight to the searches on Brainchip though. I dare say, it likely has a lot to do with retail investors like us searching for clues!
Computer engineering becomes part of inaugural program focused on neuromorphic technologies
RIT recently became one of the inaugural academic partners in the BrainChip University AI Accelerator Program. As part of the partnership, RIT’s computer engineering program will receive hardware as well as lecture modules for classes detailing how the novel chips can be programmed and used to...
Hello Wilzy... im not sure why you called my name out? Typically I just read everything here but be silent???
alwaysgreen
Top 20
alwaysgreen
Top 20
Sirod69
bavarian girl ;-)
“RISC-V products are just better. Better power, better performance, and more.”
@jackckang
summed it up perfectly at the #RISCVSummit, highlighting why we’re going to see RISC-V “take off everywhere” with SiFive’s comprehensive portfolio.
@jackckang
summed it up perfectly at the #RISCVSummit, highlighting why we’re going to see RISC-V “take off everywhere” with SiFive’s comprehensive portfolio.
Even .1 of MS would be $25....now that's a pretty healthy return from current SP. I hope that's not too much to expect in 5 yearsMaybe but it means we'll have a market cap of 180 billion USD and I'll be able to retire on a private island and bring all my friends and family along with me.
I don't need more than that
Rochester, NY
FYI
The home of the aforementioned Rochester Institute of Technology, Xerox, Eastman Kodak and the Eastman School of Music and Bausch and Lomb.
The birthplace of Susan B Anthony.....leader in women's suffrage movement. Not too far away Joseph Smith had his visions leading to the founding of the LDS church, Book of Mormon and migration to Utah etc
Remarking on the terrible weather in Rochester a classmate (who was originally from Wyoming) issued his famous (to me at least) quote....."the best view of Rochester is in the rear view mirror."
We were both enduring the weather and studies at the U. of Rochester School of Medicine.
BrainChip....thank you for inspiring a brief moment of nostalgia.
FYI
The home of the aforementioned Rochester Institute of Technology, Xerox, Eastman Kodak and the Eastman School of Music and Bausch and Lomb.
The birthplace of Susan B Anthony.....leader in women's suffrage movement. Not too far away Joseph Smith had his visions leading to the founding of the LDS church, Book of Mormon and migration to Utah etc
Remarking on the terrible weather in Rochester a classmate (who was originally from Wyoming) issued his famous (to me at least) quote....."the best view of Rochester is in the rear view mirror."
We were both enduring the weather and studies at the U. of Rochester School of Medicine.
BrainChip....thank you for inspiring a brief moment of nostalgia.
Similar threads
