BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
cosors
👀
Dr. Arpan Pal will be presenting on the following topics at the 30th IEEE International Conference on Image Processing in Kuala Lumpur on 8-11 October:
"IES #5
Title: Low Power Image Processing on Constrained Devices Using Tiny ML and Neuromorphic Approach
Speaker: Arpan Pal, TCS Research, Tata Consultancy Services, India
Abstract: More and more of IoT based intelligent systems demand embedding the analytics and AI on the edge device. This is driven by the needs of low latency, network unavailability/unreliability and a need for inherent privacy/security. However, edge devices are usually constrained in terms of compute, memory and power which poses challenges for such deployments.
In this presentation we will first introduce a Tiny Edge Wizard that can take large Deep Neural Network models and try to reduce their size / improve their latency automatically using an innovative integrated approach that uses both reduction using pruning using Lottery Ticket Hypothesis (LTH) [1] and synthesis using Neural Architecture Search (NAS) [2]. The automation helps in reducing development time, helps reducing over-parameterised models and minimises the human-expert time. We will take example use cases of medical image processing and manufacturing shop-floor inspection to demonstrate the efficacy of our proposed system.
Next we will introduce Neuromorphic Computing and Spiking Neural Networks (SNN) as a means towards ultra-low power intelligent processing at edge. We cover design and implementation of both efficient spike encoders and spiking neural models on neuromorphic chipsets. We will present application use cases around gesture recognition for human-robot interaction [3] and lossless image compression onboard nano satellites [4] and present some interesting results.
Finally we will conclude with technology trends we see in this area that also relates to sustainable analytics to negate the ever-increasing energy consumption trends in computation." Hear ye!
https://2023.ieeeicip.org/industry-expert-sessions/
@Frangipani
(Even though the article [3] probably refers to IBM)
[3] Arun M. George, Dighanchal Banerjee, Sounak Dey, Arijit Mukherjee and P. Balamurali, “A Reservoir-based Convolutional Spiking Neural Network for Gesture Recognition from DVS Input,” International Joint Conference on Neural Networks (IJCNN), 2020
[4] Sayan Kahali, Sounak Dey, Chetan S. Kadway, Arijit Mukherjee, Arpan Pal and Manan Suri, “Low-Power Lossless Image Compression on Small Satellite Edge Using Spiking Neural Network,” International Joint Conference on Neural Networks (IJCNN), 2023
I remember
here also Dr. Arpan Pal
https://thestockexchange.com.au/threads/brn-discussion-ongoing.1/post-371567
Last edited:
wilzy123
Founding Member
cosors
👀
By 'chance' I stumbled across an article in indian press. I left it at the bar counter.
A lot of students were keen on Neuromorphic Engineering.
https://thestockexchange.com.au/threads/the-brainchip-bar-friendly-banter-anything-goes-all-welcome.170934/post-374051
A lot of students were keen on Neuromorphic Engineering.
https://thestockexchange.com.au/threads/the-brainchip-bar-friendly-banter-anything-goes-all-welcome.170934/post-374051
Fullmoonfever
Top 20
Recently released Aug paper where the heading caught my eye given today's awesome discovery shared by @Learning
Unfortunately, they used Loihi 2 v CNN Xlinx but I found the statements at the end of the abstract further validation of neuromorphic and interest via the space industry.
Akida, right place, right time hopefully with players like ANT61, EDGX, Intellisense etc.
The latest satellite communication (SatCom) missions are characterized by a fully reconfigurable on-board software-defined payload, capable of adapting radio resources to the temporal and spatial variations of the system traffic. As pure optimization-based solutions have shown to be computationally tedious and to lack flexibility, machine learning (ML)-based methods have emerged as promising alternatives. We investigate the application of energy-efficient brain-inspired ML models for on-board radio resource management. Apart from software simulation, we report extensive experimental results leveraging the recently released Intel Loihi 2 chip. To benchmark the performance of the proposed model, we implement conventional convolutional neural networks (CNN) on a Xilinx Versal VCK5000, and provide a detailed comparison of accuracy, precision, recall, and energy efficiency for different traffic demands. Most notably, for relevant workloads, spiking neural networks (SNNs) implemented on Loihi 2 yield higher accuracy, while reducing power consumption by more than 100× as compared to the CNN-based reference platform. Our findings point to the significant potential of neuromorphic computing and SNNs in supporting on-board SatCom operations, paving the way for enhanced efficiency and sustainability in future SatCom systems.
Unfortunately, they used Loihi 2 v CNN Xlinx but I found the statements at the end of the abstract further validation of neuromorphic and interest via the space industry.
Akida, right place, right time hopefully with players like ANT61, EDGX, Intellisense etc.
Energy-Efficient On-Board Radio Resource Management for Satellite Communications via Neuromorphic Computing
The latest satellite communication (SatCom) missions are characterized by a fully reconfigurable on-board software-defined payload, capable of adapting radio resources to the temporal and spatial variations of the system traffic. As pure optimization-based solutions have shown to be computationally tedious and to lack flexibility, machine learning (ML)-based methods have emerged as promising alternatives. We investigate the application of energy-efficient brain-inspired ML models for on-board radio resource management. Apart from software simulation, we report extensive experimental results leveraging the recently released Intel Loihi 2 chip. To benchmark the performance of the proposed model, we implement conventional convolutional neural networks (CNN) on a Xilinx Versal VCK5000, and provide a detailed comparison of accuracy, precision, recall, and energy efficiency for different traffic demands. Most notably, for relevant workloads, spiking neural networks (SNNs) implemented on Loihi 2 yield higher accuracy, while reducing power consumption by more than 100× as compared to the CNN-based reference platform. Our findings point to the significant potential of neuromorphic computing and SNNs in supporting on-board SatCom operations, paving the way for enhanced efficiency and sustainability in future SatCom systems.
Fullmoonfever
Top 20
I see our newly found friends are presenting on their EDGX-1 in a couple of days at an ESA conference
Can't get the presso materials as locked at the mo
Hope they crush the presso...must have some confidence in it you'd think
indico.esa.int
Can't get the presso materials as locked at the mo
Hope they crush the presso...must have some confidence in it you'd think
EDHPC 2023 - European Data Handling & Data Processing Conference
EDHPC 2023The first European Data Handling & Data Processing Conference – EDHPC 2023 – was held from the 2nd to the 6th of October 2023 in Juan-Les-Pins (French Riviera). It is organised by the European Space Agency (ESA), with the support of the local tourist office. Find the latest information...
robsmark
Regular
Fullmoonfever
Top 20
EDGX definitely getting amongst it.I see our newly found friends are presenting on their EDGX-1 in a couple of days at an ESA conference
Can't get the presso materials as locked at the mo
Hope they crush the presso...must have some confidence in it you'd think
EDHPC 2023 - European Data Handling & Data Processing Conference
EDHPC 2023The first European Data Handling & Data Processing Conference – EDHPC 2023 – was held from the 2nd to the 6th of October 2023 in Juan-Les-Pins (French Riviera). It is organised by the European Space Agency (ESA), with the support of the local tourist office. Find the latest information...
View attachment 46124
After the ESA conference they are moving on to the below as an exhibitor.
Space Tech Expo Exhibitor | Space Tech Expo Europe 2023 Conference Trade Show Event & Exhibition
14 – 16 November 2023 // Bremen, Germany
EXHIBITOR LIST 2023
Booth Number: K41
Contact
WebsiteAddress:
Groeneweg 17,
Aalst,
Belgium,
9320.
EDGX
EDGX is a belgian spacetech startup pioneering the next generation of onboard AI technologies for space. We envision a future where intelligent space system bring huge economical and societal benefits for man and machine here on earth.
Categories: Design / Engineering Services || Research || Subsystem / Systems Supplier
DingoBorat
Slim
"IMO, the current stock market totally underestimates mRNA technology’s revolutionary potential"
Oh it's certainly a money spinner, I don't think anyone would argue that point
I agree, that in things I've seen, it does have great potential, in some areas, where current medicine isn't cutting the mustard.
Living in one of the most mRNA vaccinated countries in the World, I don't think I would find many, that would agree with the extremely small amount of information, that I know on the subject..
Are you aware though, that we also have some of the greatest excess deaths (above the mean) in Australia, which is over 9000 people, for the first 30 weeks of this year?
And that's not from respiratory illnesses..
Of course there is no direct provable correlation..
Hmm what's different in the World (excess deaths seeming to be high in countries with high mRNA vax rates)..
Just a curiosity I guess
..I would say that's been getting a lot of airtime in the mainstream media aye?
I mean, imagine if there were 9000 deaths above the mean "so far" this year, on the roads..
Do you agree with giving the largely untested (well it's still in progress anyway) mRNA vaccine, to people who are not in a compromised health position?
Say like kids.. That it was shown early on were little affected by the virus?..
Oh but that's to protect the vulnerable right?
Even though the drug companies didn't test whether the mRNA vaccines reduced transmissibility of the virus (which it was later found that it didn't).
It was known very early in the piece, that the virus only seriously affected very few (the vunerable).
In fact most were little affected, remember the ads?
"I just had a runny nose"
"I had a sore throat"
"Hey, I can't smell my coffee!"
"We all had Covid"..
It's true, that in various cases, that we don't "know" who the vunerable are..
So just give it to everyone, just to play safe right?
But then what about all these excess deaths
..I guess just giving it to everyone, does do one thing though, doesn't it.
Which brings my overly long post, back to it's beginning..
There are those that would call some of the information out there "mis-information" others call it the "missing-information"..
"I'm guilty as charged officer
"For lack of a better term, any "hate" mail heckling or mocking, to my inbox please
Last edited:
CrabmansFriend
Regular
I really don't want to start an off-topic discussion as discussions/views regarding covid vaccines (and mrna technology especially) are even more heated than BRN's stock price"IMO, the current stock market totally underestimates mRNA technology’s revolutionary potential"
Oh it's certainly a money spinner, I don't think anyone would argue that point
I agree, that in things I've seen, it does have great potential, in some areas, where current medicine isn't cutting the mustard.
Living in one of the most mRNA vaccinated countries in the World, I don't think I would find many, that would agree with the extremely small amount of information, that I know on the subject..
Are you aware though, that we also have some of the greatest excess deaths (above the mean) in Australia, which is over 9000 people, for the first 30 weeks of this year?
And that's not from respiratory illnesses..
Of course there is no direct provable correlation..
Hmm what's different in the World (excess deaths seeming to be high in countries with high mRNA vax rates)..
Just a curiosity I guess
I would say that's been getting a lot of airtime in the mainstream media aye?
I mean, imagine if there were 9000 deaths above the mean "so far" this year, on the roads..
Do you agree with giving the largely untested (well it's still in progress anyway) mRNA vaccine, to people who are not in a compromised health position?
Say like kids.. That it was shown early on were little affected by the virus?..
Oh but that's to protect the vulnerable right?
Even though the drug companies didn't test whether the mRNA vaccines reduced transmissibility of the virus (which it was later found that it didn't).
It was known very early in the piece, that the virus only seriously affected very few (the vunerable).
In fact most were little affected, remember the ads?
"I just had a runny nose"
"I had a sore throat"
"Hey, I can't smell my coffee!"
"We all had Covid"..
It's true, that in various cases, that we don't "know" who the vunerable are..
So just give it to everyone, just to play safe right?
But then what about all these excess deaths
I guess just giving it to everyone, does do one thing though, doesn't it.
Which brings my overly long post, back to it's beginning..
There are those that would call some of the information out there "mis-information" others call it the "missing-information"..
"I'm guilty as charged officer
View attachment 46125
But I think it often happens that anger about government measures/vaccination recommendations etc. (righteous or not) gets mixed up with knowledge/opinion about the technology (mRNA). I think a litte bit more differentiation would be a good thing sometimes.
DingoBorat
Slim
In one of the early Mercedes BrainChip related releases, they said they had developed "hardware and software solutions" with AKIDA.
Tothemoon24
Top 20
DingoBorat
Slim
Was already pretty happy with Diogenese's response.
His opening line, about not being his job to "mark essays" said it all
Google’s Pixel 8 Pro Will Herald A Brave New World Of AI Smartphones
Google will want the Pixel 8 and Pixel 8 Pro to be best-sellers, but these phones have another role to play beyond market share.
www.forbes.com
The Pixel line has been through several design iterations, with the current hardware deriving from the Pixel 6. Yet the real change has been Google switching away from the various Qualcomm Snapdragon system-on-chips to its own design. The Tensor chipset has been tailored away from outright performance towards supporting computational routines - notably for AI and Machine Learning - to happen on the device rather than be passed over to the cloud.
The use of hardware to improve AI and ML; a consistent approach across first- and second-party apps from Google and the individual manufacturers; and a long window of support and security updates. The Pixel 8 and Pixel 8 Pro will be seen as consumer-focused machines, but the long-term value is in crystallising Google’s view of mobility across the Android ecosystem.
Similar threads
