BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Pom down under
Top 20
Nothing at all about us in this, but thought Iād post it anyway
Last edited:
Jacob Kiraz
Member
O
Obvious the Reneas and Megachips deal doesn't make us rich overnight
Pom down under
Top 20
Guess you are right @DingoBorat and 300k shares is a better number than 200k only 75k to go between my personal and super totals and I better hurry up while the downrampers on here and HC are working overtime, plus Iāve gotten an extra 4000 plus shares buying them on market and not on that fab offer the company were giving us SH 
Last edited:
Soon top 50?Guess you are right @DingoBorat and 300k shares is a better number than 200k only 75k to go between my personal and super totals and I better hurry up while the downrampers on here and HC are working overtime, plus Iāve gotten an extra 4000 plus shares buying them on market and not on that fab offer the company were giving us SH
View attachment 67764
Pom down under
Top 20
Depends how long the shorters carry on playing the game and also it depends on the if we get no news in the next 4-6 months as Iām completely out of money as I just spent my airfare to the uk on some more today
plus Iām going for a takeover instead as top 50 sounds boring.
Last edited:
Evening Chippers,
Listening to the ABC radio ... new technology using Ai to take photos of the placenta once child is born ... looks for anything which may be amiss.
Company called PlacentaVision.
Also of note , University of Pennsylvania helping on the technical side.
* whilst getting lost in all of this ... also came appon this site.
* for the technically minded only, got a light nose bleed on quick perusal.
* this info may have been shared by others , those in the know will know if it's new or not.
Regards,
Esq.
Listening to the ABC radio ... new technology using Ai to take photos of the placenta once child is born ... looks for anything which may be amiss.
Company called PlacentaVision.
Also of note , University of Pennsylvania helping on the technical side.
* whilst getting lost in all of this ... also came appon this site.
* for the technically minded only, got a light nose bleed on quick perusal.
* this info may have been shared by others , those in the know will know if it's new or not.
Regards,
Esq.
Pom down under
Top 20
Weird as most people have there photos taken with the baby.
Tata Elxsi on LinkedIn: The Road to Innovation: Unlocking the Potential of Software-Defined Vehicle
This Friday, on the 'Tata Elxsi' YouTube channel, we feature an exclusive interview where Ramprasshanth Viswanathan (Product Manager, Connected Digitalā¦
Fullmoonfever
Top 20
Nice 
Paper HERE
Submitted on 21 Jul 2024]
Objective: This work explores use of a few-shot transfer learning method to train and implement a convolutional spiking neural network (CSNN) on a BrainChip Akida AKD1000 neuromorphic system-on-chip for developing individual-level, instead of traditionally used group-level, models using electroencephalographic data. The efficacy of the method is studied on an advanced driver assist system related task of predicting braking intention. Main Results: Efficacy of the above methodology to develop individual specific braking intention predictive models by rapidly adapting the group-level model in as few as three training epochs while achieving at least 90% accuracy, true positive rate and true negative rate is presented. Further, results show an energy reduction of over 97% with only a 1.3x increase in latency when using the Akida AKD1000 processor for network inference compared to an Intel Xeon CPU. Similar results were obtained in a subsequent ablation study using a subset of five out of 19 channels.
Significance: Especially relevant to real-time applications, this work presents an energy-efficient, few-shot transfer learning method that is implemented on a neuromorphic processor capable of training a CSNN as new data becomes available, operating conditions change, or to customize group-level models to yield personalized models unique to each individual.
5. Conclusion
The results show that the methodology presented was effective to develop individual-level models deployed on a state-of-the-art neuromorphic processor with predictive abilities for ADAS relevant tasks, specifically braking intent detection.
This study explored a novel application of deep SNNs to the field of ADAS using a neuromorphic processor by creating and validating individual-level braking intent classification models with data from three experiments involving pseudo-realistic conditions. These conditions included cognitive atrophy through physical fatigue and real-time distraction
and providing braking imperatives via commonly encountered visual stimulus of traffic lights. The method presented demonstrates that individual-level models could be quickly created with a small amount of data, achieving greater than 90% scores across all three classification performance metrics in a few shots (three epochs) on average for both the ACS and FCAS. This demonstrated the efficacy of the method for different participants operating under non-ideal conditions and using realistic driving cues and further suggests that a reduced data acquisition scheme might be feasible in the field.
Furthermore, the applicability to energy-constrained systems was demonstrated through comparison of the inference energy consumed with a very powerful CPU in which the Akida processor offered energy savings of 97% or greater. The Akida processor was also shown to be competitive in inference latency compared to the CPU. Future work could
include implementation of the method presented on a larger number of participants, other neuromorphic hardware, different driving scenarios, and in real-world scenarios where individual-level models are created by refining previously developed group-level models in real time.
Paper HERE
Submitted on 21 Jul 2024]
Few-Shot Transfer Learning for Individualized Braking Intent Detection on Neuromorphic Hardware
Nathan Lutes, Venkata Sriram Siddhardh Nadendla, K. KrishnamurthyObjective: This work explores use of a few-shot transfer learning method to train and implement a convolutional spiking neural network (CSNN) on a BrainChip Akida AKD1000 neuromorphic system-on-chip for developing individual-level, instead of traditionally used group-level, models using electroencephalographic data. The efficacy of the method is studied on an advanced driver assist system related task of predicting braking intention. Main Results: Efficacy of the above methodology to develop individual specific braking intention predictive models by rapidly adapting the group-level model in as few as three training epochs while achieving at least 90% accuracy, true positive rate and true negative rate is presented. Further, results show an energy reduction of over 97% with only a 1.3x increase in latency when using the Akida AKD1000 processor for network inference compared to an Intel Xeon CPU. Similar results were obtained in a subsequent ablation study using a subset of five out of 19 channels.
Significance: Especially relevant to real-time applications, this work presents an energy-efficient, few-shot transfer learning method that is implemented on a neuromorphic processor capable of training a CSNN as new data becomes available, operating conditions change, or to customize group-level models to yield personalized models unique to each individual.
5. Conclusion
The results show that the methodology presented was effective to develop individual-level models deployed on a state-of-the-art neuromorphic processor with predictive abilities for ADAS relevant tasks, specifically braking intent detection.
This study explored a novel application of deep SNNs to the field of ADAS using a neuromorphic processor by creating and validating individual-level braking intent classification models with data from three experiments involving pseudo-realistic conditions. These conditions included cognitive atrophy through physical fatigue and real-time distraction
and providing braking imperatives via commonly encountered visual stimulus of traffic lights. The method presented demonstrates that individual-level models could be quickly created with a small amount of data, achieving greater than 90% scores across all three classification performance metrics in a few shots (three epochs) on average for both the ACS and FCAS. This demonstrated the efficacy of the method for different participants operating under non-ideal conditions and using realistic driving cues and further suggests that a reduced data acquisition scheme might be feasible in the field.
Furthermore, the applicability to energy-constrained systems was demonstrated through comparison of the inference energy consumed with a very powerful CPU in which the Akida processor offered energy savings of 97% or greater. The Akida processor was also shown to be competitive in inference latency compared to the CPU. Future work could
include implementation of the method presented on a larger number of participants, other neuromorphic hardware, different driving scenarios, and in real-world scenarios where individual-level models are created by refining previously developed group-level models in real time.
Frangipani
Top 20
Nice
Paper HERE
Submitted on 21 Jul 2024]
Few-Shot Transfer Learning for Individualized Braking Intent Detection on Neuromorphic Hardware
Nathan Lutes, Venkata Sriram Siddhardh Nadendla, K. Krishnamurthy
Objective: This work explores use of a few-shot transfer learning method to train and implement a convolutional spiking neural network (CSNN) on a BrainChip Akida AKD1000 neuromorphic system-on-chip for developing individual-level, instead of traditionally used group-level, models using electroencephalographic data. The efficacy of the method is studied on an advanced driver assist system related task of predicting braking intention. Main Results: Efficacy of the above methodology to develop individual specific braking intention predictive models by rapidly adapting the group-level model in as few as three training epochs while achieving at least 90% accuracy, true positive rate and true negative rate is presented. Further, results show an energy reduction of over 97% with only a 1.3x increase in latency when using the Akida AKD1000 processor for network inference compared to an Intel Xeon CPU. Similar results were obtained in a subsequent ablation study using a subset of five out of 19 channels.
Significance: Especially relevant to real-time applications, this work presents an energy-efficient, few-shot transfer learning method that is implemented on a neuromorphic processor capable of training a CSNN as new data becomes available, operating conditions change, or to customize group-level models to yield personalized models unique to each individual.
5. Conclusion
The results show that the methodology presented was effective to develop individual-level models deployed on a state-of-the-art neuromorphic processor with predictive abilities for ADAS relevant tasks, specifically braking intent detection.
This study explored a novel application of deep SNNs to the field of ADAS using a neuromorphic processor by creating and validating individual-level braking intent classification models with data from three experiments involving pseudo-realistic conditions. These conditions included cognitive atrophy through physical fatigue and real-time distraction
and providing braking imperatives via commonly encountered visual stimulus of traffic lights. The method presented demonstrates that individual-level models could be quickly created with a small amount of data, achieving greater than 90% scores across all three classification performance metrics in a few shots (three epochs) on average for both the ACS and FCAS. This demonstrated the efficacy of the method for different participants operating under non-ideal conditions and using realistic driving cues and further suggests that a reduced data acquisition scheme might be feasible in the field.
Furthermore, the applicability to energy-constrained systems was demonstrated through comparison of the inference energy consumed with a very powerful CPU in which the Akida processor offered energy savings of 97% or greater. The Akida processor was also shown to be competitive in inference latency compared to the CPU. Future work could
include implementation of the method presented on a larger number of participants, other neuromorphic hardware, different driving scenarios, and in real-world scenarios where individual-level models are created by refining previously developed group-level models in real time.
Hi Fullmoonfever,
was just about to post a link to that paper, too - it goes along with the webinar I posted about a month ago:
The paperās first author is Nathan Lutes, a PhD student at Missouri University of Science and Technology (graduating in December), who has also been working as a Guidance Navigation and Control Engineer for Boeing since 2021.
Space Cadet
Regular
Analyzing Amazon SageMaker and BrainChip Akida Neuromorphic Processor, pioneering approaches in machine learning
Can someone find this and post please. Having trouble with phone. Put it down to user error.
It's fairly new from May.
TIA
SC
Can someone find this and post please. Having trouble with phone. Put it down to user error.
It's fairly new from May.
TIA
SC
Space Cadet
Regular
I say WORM .
Frangipani
Top 20
Thanks to Lawrence Livermore National Laboratoryās summer intern Henry Gleason, a B.Sc. Aerospace Engineering student at Purdue University, we now know that LLNL researchers are looking at Akida 2.0 / TENNs for future satellite imagery processingā¦ š° 
www.linkedin.com
During his summer internship at Lawrence Livermore National Laboratory, Henry Gleason collaborated with the following two gentlemen on his proof-of-concept Self-Tasking Satellite Network Model: Shervin Kiannejad, an Embedded System Engineer at LLNL āworking on some of the most cutting edge technological projects out thereā (his main interest being RF), and George Dankiewicz, who has been an Engineering Manager with Lockheed Martin for more than 40 yearsā¦
I am excited to announce that I have completed my first internship this summer in Lawrence Livermore National Laboratory's Space Science and Security Program within the Global Security Directorate. | Henry Gleason
I am excited to announce that I have completed my first internship this summer in Lawrence Livermore National Laboratory's Space Science and Security Program within the Global Security Directorate. Over the last 13 weeks, I have been doing research and designing models for a Global Satellite...
During his summer internship at Lawrence Livermore National Laboratory, Henry Gleason collaborated with the following two gentlemen on his proof-of-concept Self-Tasking Satellite Network Model: Shervin Kiannejad, an Embedded System Engineer at LLNL āworking on some of the most cutting edge technological projects out thereā (his main interest being RF), and George Dankiewicz, who has been an Engineering Manager with Lockheed Martin for more than 40 yearsā¦
Last edited:
Dijon101
Regular
Depends how long the shorters carry on playing the game and also it depends on the if we get no news in the next 4-6 months as Iām completely out of money as I just spent my airfare to the uk on some more today
So expecting crypto to go parabolic in the next 4-6 months and hoping to transfer profits over to BRN ???
(me too !! I'm out of funds as well!!)
Pom down under
Top 20
Only holding XRP since last year and got in at 0.70au cashed a little out and holding long term hoping for a miracle along with BRN
rgupta
Regular
How it was a news now, the information was all there in last 4cGood morning fellow chippers!
What a great news to wake up to.
New CMO. Looks like big changes are being made! Great move Sean and co!
OMG how exciting! This will surely work for us
Can't help to think that if I recall correctly, our previous CMO Nandan had a lot of experience. He was Ex-ARM and came from Amazon. He was touted as the one we need to drive our product marketing globally through his connections. This was 2 years ago and got a lot of share holders excited. Then he left quietly.
I'd give this new guy 2-3 years top.
GLTAH
Not advice
Similar threads
