BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
equanimous
Norse clairvoyant shapeshifter goddess
Interesting that this research scientist from Brainchip presented this Paper in China.
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
View attachment 40315
Harshil Patel on LinkedIn: #aicas2023 #ieee #edgeai #neuromorphic #brainchip
Last month I had the pleasure of presenting our paper titled "Bringing Touch to the Edge: a Neuromorphic Processing Approach for Event-Based Tactile Systems"…www.linkedin.com
There is a ton of research papers from China on SNN so curious as to where this is heading..
How stupid is market to dump this amazing stock with no revenue to 30 cents level? Market is now suddenly capable of valuing this company & announcements? Wowww amazinggg
Although I shouldn't complain about green, they have been pump and dumping on the fluffy announcements...
Last edited:
Tothemoon24
Top 20
Australian AI chipmaker rises on US patent grant
By Staff Writer on Jul 19, 2023 4:54PM
BrainChip gets more backing.
Australia's BrainChip Holdings surged nearly 11 percent to mark its best day in six weeks after the advanced artificial intelligence chipmaker received a patent grant in the United States on one of its neuromorphic processors.Neuromorphic processors mimic the way the brain works and allow devices to use signals from physical neurons to carry out computations.
Shares of BrainChip were trading 7.4 percent higher at A$0.3975, their highest level since June 19, and were the biggest gainers on the ASX 200 benchmark index.
More than 11.3 million shares had changed hands, nearly double the 30-day average of around 6.4 million.
The patent protects the neural processor's feature of reconfigurability, which "provides a significant advantage, enabling the development of multipurpose and cost-effective hardware designs", BrainChip said.
BrainChip now has 17 issued patents, with around 30 patent applications pending in the United States, Europe, Australia, Canada, Japan, Korea, India, among others.
Tothemoon24
Top 20
BrainChip Further Bolsters its Patent Portfolio with a Ground-breaking Edge Learning Technology Patent (ASX:BRN)
19th July 2023 | Technology News | StockHoundBrainChip Holdings Ltd (ASX: BRN, OTCQX: BRCHF, ADR: BCHPY), the commercial provider of ultra-low power neuromorphic AI IP, has announced the addition of a strategic patent to its growing portfolio. The new patent solidifies the company’s competitive research strength in the rapidly evolving AI field.
The patent, US 11,704,549, entitled “Event-Based Classification of Features in a Reconfigurable and Temporally Coded Convolutional Spiking Neural Network” was issued on July 18, 2023, by the United States Patent and Trademark Office (USPTO). BrainChip views this patent as a significant intellectual property (IP) asset that fortifies the company’s unique approach to learning on a neuromorphic chip or system.
The patent principally shields BrainChip’s innovative neuromorphic processor, a unique configuration that performs convolutions on digital input data after converting it into spikes. The inclusion of this new patent extends the scope of protection already provided to BrainChip in one of its earlier patents, hence further amplifying the commercial value of their patent portfolio.
Moreover, the patent safeguards the functionality of reconfigurability in a neural processor. This feature allows the processor to function either as a convolutional neural processor or a fully connected neural processor, offering a significant advantage by facilitating the creation of versatile and cost-effective hardware designs like the Akida.
With the issuance of the US 11,704,549, BrainChip’s portfolio now boasts 17 patents (12 in the US, 3 in Australia, and one each in Europe and China). The company also has approximately 30 patent applications currently under review in several jurisdictions, including the US, Europe, Australia, Canada, Japan, Korea, India, Brazil, Russia, Mexico, and Israel.
BrainChip is renowned globally for its work in edge AI on-chip processing and learning. Their neuromorphic processor, Akida, mimics the human brain to analyse essential sensor inputs at the point of acquisition, processing data with efficiency, precision, and energy conservation. By keeping machine learning local to the chip and independent of the cloud, BrainChip reduces latency while enhancing privacy and data security. The company is demonstrating that on-chip AI, located close to the sensor, is the potential future of real-world applications like connected cars, consumer electronics, and industrial IoT.
Damo4
Regular
BRN or BCF?
Bait everywhere
How much are you selling them for? I might be interestedThanks for the heads up. Selling all my shares now
Not advice.
The announcement needs to be taken in the context of exponential growth predictions for AI at the Edge.
The strong patent portfolio on its own could well be worth many times the current market cap.
That is my safety net and why i never panic over the day to day noise of the market in relation to BRN.
The strong patent portfolio on its own could well be worth many times the current market cap.
That is my safety net and why i never panic over the day to day noise of the market in relation to BRN.
The true value of our continuing list of granted patents still hasn't been fully understood by many market insiders, these
roadblocks will eventually send all suitors to Brainchip's door, because to advance down this track that Brainchip has been on
for over 20 years, taking Peters research into consideration, will require a password and a hell of a lot of money for a company
to either join us or for us to pull aside and let them/it take the reins. That day will come I'm sure of it, but for now, I personally
believe that our patents are very much undervalued due to a lack of real understanding as to the power that they command.
Patents don't just fall out of a Corn Flakes box, now do they.
Tech
roadblocks will eventually send all suitors to Brainchip's door, because to advance down this track that Brainchip has been on
for over 20 years, taking Peters research into consideration, will require a password and a hell of a lot of money for a company
to either join us or for us to pull aside and let them/it take the reins. That day will come I'm sure of it, but for now, I personally
believe that our patents are very much undervalued due to a lack of real understanding as to the power that they command.
Patents don't just fall out of a Corn Flakes box, now do they.
Tech
Tothemoon24
Top 20
Cadence, GlobalFoundries, Hoerzentrum Oldenburg and Leibniz University Hannover Collaborate to Advance Hearing Aid Technology
18 Jul 2023
Highlights:
- Cadence Tensilica Fusion G6 DSP and Xtensa LX7 processor, the Cadence digital full flow and GF’s 22FDX platform enable first-of-its-kind binaural hearing aid SoC prototype
- New SmartHeAP SoC prototype lets customers create hearing aids that process sound more optimally and reduce background noise
The SmartHeAP SoC prototype provides hearing aid companies with all the components required to create and reprogram hearing devices that improve a wearer’s hearing experience. Some of the key benefits the SmartHeAP SoC prototype provides include:
- Binaural hearing technology: Hearing aids in the right and left ears communicate with one another, enabling a wearer to pick up sounds from the full auditory scene without destroying the binaural cues.
- Improved hearing loss compensation capabilities: Through advanced algorithms, the SoC automatically analyzes the incoming signal and provides adaptive sound amplification that is customized to the wearer’s unique hearing needs.
- High processing capacity with minimal power consumption: Provides the hearing aid wearer with optimal sound quality in real time while conserving power and extending hearing aid battery life.
- Cost Savings: The hearing aid software can be quickly upgraded without replacing the hardware, saving both the wearers and the hearing aid companies money.
- Faster time to market: Hearing aid companies of all sizes can effectively compete in the market due to the ease of use of high-level programming languages, which enable faster innovation cycles.
“By collaborating with GF, Hoerzentrum Oldenburg gGmbH and Leibniz University Hannover on the SmartHeAP SoC prototype, our research and technology advancements are improving the hearing experience,” said Rishi Chugh, vice president, product management in the IP Group at Cadence. “Each organization involved brought a unique perspective to the project, and delivering a successful prototype that can make a positive impact on the hearing experience is the ultimate reward. The Cadence digital full flow and Tensilica Fusion G6 DSP and Xtensa LX7 controller were foundational for this project, and the consortium is already seeing a high level of interest in the prototype from the hearing aid industry.”
The Tensilica Fusion G6 DSP and Xtensa LX7 processor, as well as the digital full flow, support Cadence’s broader Intelligent System Design™ Strategy, enabling SoC design excellence. To learn more about the Cadence Tensilica IP, please visit www.cadence.com/go/tensilicash, and to learn more about the Cadence digital full flow, visit www.cadence.com/go/digitalffsh. To learn about the GF 22FDX platform, go to gf.com/technology-platforms. For more information on the full SmartHeAP system, visit https://www.elektronikforschung.de/projekte/smartheap.
Consortium Quotes
“The collaboration between GF, Cadence, Hoerzentrum Oldenburg gGmbH and Leibniz University Hannover resulted in a successful hearing aid SoC that consumes significantly less power while providing hearing aid wearers with the first-of-its-kind binaural technology that transforms how they process a complete auditory scene. By harnessing the power of the Cadence Tensilica DSP and digital full flow combined with the energy efficiency of the 22FDX platform, we delivered compelling PPA results.”
- Ted Letavic, corporate fellow, GlobalFoundries
“Hoerzentrum Oldenburg gGmbH stands for a unique combination of interdisciplinary, demand-oriented hearing research, customer-oriented services and market-oriented product developments. It has successfully developed numerous innovative methods into medical products and brought them to market. Through our current collaboration with Cadence, GF and Leibniz University Hannover, we’ve succeeded in advancing relevant research for the improvement and customization of hearing aid technology and contributing to the development of advanced algorithms.”
- Dr. Kamil Adiloğlu, research engineer, Hoerzentrum Oldenburg gGmbH
“The Institute for Microelectronic Systems (IMS) at Leibniz University Hannover is focused on the design and implementation of digital and analog circuits and systems as well as in the development of tools for the automation of the design process. IMS has long-standing expertise in mapping powerful and compute-intensive algorithms to Cadence’s Tensilica architecture, using the full potential of its instruction set extensibility. By collaborating with Cadence, GF and Hoerzentrum Oldenburg gGmbH, we’re bringing the joy of hearing back to hearing aid wearers with this new SoC prototype.”
- Prof. Holger Blume, managing director, Institute for Microelectronic Systems, Leibniz University Hannover
About Cadence
Cadence is a pivotal leader in electronic systems design, building upon more than 30 years of computational software expertise. The company applies its underlying Intelligent System Design strategy to deliver software, hardware and IP that turn design concepts into reality. Cadence customers are the world’s most innovative companies, delivering extraordinary electronic products from chips to boards to complete systems for the most dynamic market applications, including hyperscale computing, 5G communications, automotive, mobile, aerospace, consumer, industrial and healthcare. For eight years in a row, Fortune magazine has named Cadence one of the 100 Best Companies to Work For. Learn more at cadence.com.
wilzy123
Founding Member
I don't think you're really a good card player.
FlipDollar
Never dog the boys
Strategic Patenting
Strategic patenting for many means creating and claiming intellectual property in targeted areas that are intended to create future industry recognized value. The early leaders in this field were J…
businessinnovationmanagement.com
David holland
Regular
It builds up then back to 30 cents on report
- Binaural hearing technology: Hearing aids in the right and left ears communicate with one another, enabling a wearer to pick up sounds from the full auditory scene without destroying the binaural cues.
equanimous
Norse clairvoyant shapeshifter goddess
I changed my mind.How much are you selling them for? I might be interested
Might let you borrow them for a fee, as long you give them back later
NoooooooooI changed my mind.
Might let you borrow them for a fee, as long you give them back later
Tothemoon24
Top 20
Dio ,
Can you let me know about the wires here pls
wilzy123
Founding Member
I changed my mind.
Might let you borrow them for a fee, as long you give them back later
Tothemoon24
Top 20
Qualcomm and Meta will enable the social networking company’s new large language model, Llama 2, to run on Qualcomm chips on phones and PCs starting in 2024, the companies announced today.
So far, LLMs have primarily run in large server farms, on Nvidia graphics processors, due to the technology’s vast needs for computational power and data, boosting Nvidia stock, which is up more than 220% this year. But the AI boom has largely missed the companies that make leading edge processors for phones and PCs, like Qualcomm. Its stock is up about 10% so far in 2023, trailing the NASDAQ’s gain of 36%.
The announcement on Tuesday suggests that Qualcomm wants to position its processors as well-suited for A.I. but “on the edge,” or on a device, instead of “in the cloud.” If large language models can run on phones instead of in large data centers, it could push down the significant cost of running A.I. models, and could lead to better and faster voice assistants and other apps.
Qualcomm will make Meta’s open-source Llama 2 models available on Qualcomm devices, which it believes will enable applications like intelligent virtual assistants. Meta’s Llama 2 can do many of the same things as ChatGPT, but it can be packaged in a smaller program, which allows it to run on a phone.
Qualcomm’s chips include a “tensor processor unit,” or TPU, that is well-suited for the kinds of calculations that A.I. models require. However, the amount of processing power that is available on a mobile device pales in comparison to a data center stocked with cutting-edge GPUs.
Meta’s Llama is notable because Meta published its “weights,” a set of numbers that helps govern how a particular AI model works. Doing this will allow researchers and eventually commercial enterprises to use the AI models on their own computers without asking permission or paying. Other notable LLMs, like OpenAI’s GPT-4, or Google’s Bard, are closed-source, and their weights are closely held secrets.
Qualcomm has worked with Meta closely in the past, notably on chips for its Quest virtual reality devices. It has also demoed some A.I. models running slowly on its chips, such as the open source image generator Stable Diffusion.
So far, LLMs have primarily run in large server farms, on Nvidia graphics processors, due to the technology’s vast needs for computational power and data, boosting Nvidia stock, which is up more than 220% this year. But the AI boom has largely missed the companies that make leading edge processors for phones and PCs, like Qualcomm. Its stock is up about 10% so far in 2023, trailing the NASDAQ’s gain of 36%.
The announcement on Tuesday suggests that Qualcomm wants to position its processors as well-suited for A.I. but “on the edge,” or on a device, instead of “in the cloud.” If large language models can run on phones instead of in large data centers, it could push down the significant cost of running A.I. models, and could lead to better and faster voice assistants and other apps.
Qualcomm will make Meta’s open-source Llama 2 models available on Qualcomm devices, which it believes will enable applications like intelligent virtual assistants. Meta’s Llama 2 can do many of the same things as ChatGPT, but it can be packaged in a smaller program, which allows it to run on a phone.
Qualcomm’s chips include a “tensor processor unit,” or TPU, that is well-suited for the kinds of calculations that A.I. models require. However, the amount of processing power that is available on a mobile device pales in comparison to a data center stocked with cutting-edge GPUs.
Meta’s Llama is notable because Meta published its “weights,” a set of numbers that helps govern how a particular AI model works. Doing this will allow researchers and eventually commercial enterprises to use the AI models on their own computers without asking permission or paying. Other notable LLMs, like OpenAI’s GPT-4, or Google’s Bard, are closed-source, and their weights are closely held secrets.
Qualcomm has worked with Meta closely in the past, notably on chips for its Quest virtual reality devices. It has also demoed some A.I. models running slowly on its chips, such as the open source image generator Stable Diffusion.
Similar threads
