BRN - Mercedes

Ian

Founding Member
 
  • Like
  • Fire
  • Love
Reactions: 18 users

equanimous

Norse clairvoyant shapeshifter goddess
Convolutional Spiking Neural Networks for Detecting Anticipatory Brain Potentials Using Electroencephalogram

 
  • Like
  • Thinking
Reactions: 7 users

equanimous

Norse clairvoyant shapeshifter goddess
Convolutional Spiking Neural Networks for Detecting Anticipatory Brain Potentials Using Electroencephalogram


Now who do we know who has a Convolutional Spiking Neural Network chip already commercially available:

View attachment 14268



VI. CONCLUSION AND FUTURE WORK This paper explored the use of a CSNN as a classifier for detecting features in EEG data that predict braking intention, which occurs before the actual physical activity. The EEG data for the classification experiment was collected via an in-house experiment using a pseudo-realistic testbed with the participant operating a remote-controlled vehicle using a live video feed. The CSNN performance was compared to a standard CNN and three GNN models using a 10-fold cross-validation scheme with the CSNN achieving the highest performance and with more consistency. In addition, the effect of converting the floating-point EEG data into spike trains prior to training the CSNN was studied. The best results were obtained using a threshold of 0.5, which were similar to those obtained using floating-point data, suggesting that spike train transformation might be possible with acceptable levels of performance degradation. Future work includes assessing the performance of the CSNN in different environments, particularly when the cognitive functions of the participants are stressed because of, for example, fatigue or distractions. In addition, the ability of the CSNN to decode the participant’s intention in other EEG control signals in BCI applications, such as P300, motor imagery, motor-related cortical potentials and steady-state evoked potentials would be of interest. Implementation of the CSNN on a neuromorphic platform to study energy efficiency is another area of future research.

 
  • Like
  • Fire
  • Love
Reactions: 12 users
D

Deleted member 118

Guest
  • Like
Reactions: 2 users
D

Deleted member 118

Guest
 
  • Like
Reactions: 5 users

2B3DFE3B-E408-4735-88D6-08FF2903A8D7.jpeg
 
  • Like
  • Fire
Reactions: 14 users
Second Gen already, exciting, things moving quickly in tech, and the SUV no less !!
 
  • Like
Reactions: 10 users

Terroni2105

Founding Member
As part of its first Automotive Investor Day today, Qualcomm announced a pair of new partnerships with Mercedes-Benz and Red Hat.

The latest announcements are focused on the software defined vehicle area. Mercedes-Benz will be adopting Qualcomm Snapdragon cockpit chips to power next-generation infotainment and connectivity systems. This marks a shift away from Nvidia which Mercedes has used in its current MBUX system. The first Mercedes vehicles with Qualcomm digital cockpit will launch in 2023.


i watched the Qualcomm Auto investor day posted in the Qualcomm thread, where Qualcomm CEO states they have been working with Mercedes for 3 years.
 
  • Like
  • Fire
Reactions: 12 users

toasty

Regular
As part of its first Automotive Investor Day today, Qualcomm announced a pair of new partnerships with Mercedes-Benz and Red Hat.

The latest announcements are focused on the software defined vehicle area. Mercedes-Benz will be adopting Qualcomm Snapdragon cockpit chips to power next-generation infotainment and connectivity systems. This marks a shift away from Nvidia which Mercedes has used in its current MBUX system. The first Mercedes vehicles with Qualcomm digital cockpit will launch in 2023.


i watched the Qualcomm Auto investor day posted in the Qualcomm thread, where Qualcomm CEO states they have been working with Mercedes for 3 years.

To me this is sufficient information to deduce that we are now part of the Qualcomm world..........

My opinion only
 
  • Like
Reactions: 9 users

Build-it

Regular
Since Mercedes is being discussed in the BRN Discussion thread, here's a link revive the earlier Mercedes thread :

cosors

Regular​

I just had a quick look at the new EQS. The Akida also takes control of parts of the MBUX and display controls. Do I understand this correctly?

"There is much to discover in the digital interior of the new EQS SUV: For example, the optional MBUX Hyperscreen, whose displays self-learn to adapt to your habits, and the large head-up display with augmented reality."

Dio,
Not sure if this answers your question however I believe where in the drivers seat so to speak.


And this line is interesting also..

There will soon be a new pinnacle of electric luxury from Mercedes-Benz: the made-in-America EQS SUV.

Edge Compute.
 
  • Like
  • Fire
Reactions: 9 users

Build-it

Regular
Dio,
Not sure if this answers your question however I believe where in the drivers seat so to speak.


And this line is interesting also..

There will soon be a new pinnacle of electric luxury from Mercedes-Benz: the made-in-America EQS SUV.

Edge Compute.


A little digging into MB in the USA turned up some interesting info i was unaware of considering the current macro environment concerns for Europe manufacturering the lights should stay on in America.
 

Attachments

  • Screenshot_20221002-110001_Samsung Internet.jpg
    Screenshot_20221002-110001_Samsung Internet.jpg
    915.6 KB · Views: 125
  • Screenshot_20221002-105842_Samsung Internet.jpg
    Screenshot_20221002-105842_Samsung Internet.jpg
    567.1 KB · Views: 125
  • Like
  • Fire
  • Wow
Reactions: 12 users

Pappagolla

Regular
  • Like
  • Fire
Reactions: 16 users

stuart888

Regular
Sure hope all 1000 eyes at least knows about Luminar, the Orlando Florida Lidar company.

I have a strong suspicion that the Brainchip implementation-team focused on Mercedes knows them well. 💒

Just akida dot collecting!

https://group.mercedes-benz.com/inn...omous-driving/articlepartnership-luminar.html

1673291627584.png



Why Orlando?​

There are many reasons why Luminar, the leading autonomous vehicle and lidar technology company for consumer cars and trucking, chooses Orlando as its headquarters, according to Luminar Chief Business Officer Scott Faris.

The primary one, however, is that the largest Department of Defense laser programs (Lockheed Martin, Northrop Grumman and L3 Communications) are all based in Orlando.

The region’s history in aerospace and defense has developed unparalleled expertise in the region for photonics technology. And the ability to leverage this powerful ecosystem allows Luminar to set the bar for high performance sensor technology that will ultimately make autonomous mobility safe and ubiquitous.

 
  • Like
  • Fire
  • Love
Reactions: 20 users

BaconLover

Founding Member
 
  • Love
  • Like
  • Fire
Reactions: 14 users

stuart888

Regular
Since Peter asked for some questions, just thinking. First, I would as a shareholder never surprise the BRN team with a question at the AGM or anywhere.

I would send it over and give them a heads up. They might not deem it the right time/place. They know stuff we do not, contracts being negotiated for instance. I am team Brainchip, as a stockholder.

If they let me ask a Question, I would like to be a setup for a winning answer:

Your 4-Bits Are Enough and other publishing's have highlighted your energy inference/ml efficicency. Is the industry in agreement enough where you can proclaim specific features like "keyword spotting" as Category Leading?

As a shareholder, I am not going to surprise Peter with any Linkedin questions, that he is not aware of. Again, from me, Team Brainchip.

I would love to set him up to embellish Akida versions now and upcoming in various Use Cases.
 
  • Like
  • Fire
Reactions: 6 users
Since Peter asked for some questions, just thinking. First, I would as a shareholder never surprise the BRN team with a question at the AGM or anywhere.

I would send it over and give them a heads up. They might not deem it the right time/place. They know stuff we do not, contracts being negotiated for instance. I am team Brainchip, as a stockholder.

If they let me ask a Question, I would like to be a setup for a winning answer:

Your 4-Bits Are Enough and other publishing's have highlighted your energy inference/ml efficicency. Is the industry in agreement enough where you can proclaim specific features like "keyword spotting" as Category Leading?

As a shareholder, I am not going to surprise Peter with any Linkedin questions, that he is not aware of. Again, from me, Team Brainchip.

I would love to set him up to embellish Akida versions now and upcoming in various Use Cases.
"If they let me ask a Question, I would like to be a setup for a winning answer:"
We call that a Dorothy Dixer.
Honestly I'm not a fan of them.
 
  • Like
Reactions: 3 users

Learning

Learning to the Top 🕵‍♂️
  • Like
  • Love
Reactions: 20 users

buena suerte :-)

BOB Bank of Brainchip
  • Like
  • Love
Reactions: 8 users

Bravo

If ARM was an arm, BRN would be its biceps💪!
  • Like
  • Love
  • Fire
Reactions: 13 users

Bravo

If ARM was an arm, BRN would be its biceps💪!

Bangalore Neuromorphic Engineering Workshop​

Top-tier training awaits the next generation of researchers. Inspired by Telluride (est. 1993) and Capo Caccia (est. 2007), we aim to expand their legacy into Asia-Pacific. Our objective: equip individuals with essential neuromorphic engineering skills and foster global advancement.

Date: Jan 5 - Jan 15, 2025​

Venue: Indian Institute of Science, Bangalore​


About Bangalore Neuromorphic Engineering Workshop​

Aim and Vision​

The Bangalore Neuromorphic Engineering Workshop is designed to provide first-rate training to upcoming researchers in neuromorphic technologies, building upon the legacy of influential workshops like Telluride and Capo Caccia, these workshops have significantly contributed to the development of vibrant research communities in the US and Europe. Our goal is to replicate this success in the Asia-Pacific region. We aim to equip participants with advanced skills in neuromorphic engineering and to enhance the growth of this critical field on an international scale.

Neuromorphic Engineering Origins​

Originating from the innovative work of Carver Mead and Misha Mahowald at Caltech in the 1990s, neuromorphic started with creating brain-like models of neural computation and sensing using traditional VLSI CMOS technologies. The field has expanded to include studies on new memory technologies, development of large-scale spiking neural network chips for advanced computing, and exploring theories of plasticity, learning, and both artificial and natural intelligence.

Industry and Academic Impact​

Today, the field of neuromorphic engineering is experiencing widespread adoption by major corporations such as Samsung, Sony, Intel, and IBM, alongside significant support from governmental funding agencies. Recognized as a promising avenue for developing AI computing technologies, neuromorphic approaches offer potential solutions to the inherent limitations and critical challenges faced by conventional computing systems and deep learning methodologies. By leveraging principles inspired by the brain's architecture, neuromorphic engineering endeavors to revolutionize the landscape of artificial intelligence, paving the way for more efficient, adaptive, and intelligent computing systems.

Topic Areas of BNEW-2025​

Get ready for an immersive experience at the Bangalore Neuromorphic Engineering Workshop, where you'll engage in lively discussions, hands-on sessions, and much more, exploring cutting-edge topics like:

Neuromorphic Integrated Circuits (NIC25)​

Exploring novel neuromorphic architectures, circuits, and hardware that could demonstrate neuromorphic performance advantages when compared to conventional computing architectures like central processing units (CPUs), graphical processing units (GPUs).

Topic leaders - Shantanu Chakrabartty (Washington University St. Louis), Chetan Singh Thakur (IISc)
Read More

Event-based Sensing (EBS25)​

Explore the latest developments in neuromorphic sensors that mimic the human senses, including dynamic vision sensors, SCAMP, neuromorphic radar, olfactory sensors, and more. These innovative sensors are transforming industries and applications by providing more efficient, adaptive, and human-like sensing capabilities, enabling new possibilities for robotics, autonomous systems, healthcare, and the Internet of Things (IoT).

Topic leaders - Piotr Dudek (University of Manchester), Gregory Cohen (Western Sydney University)
Read More

Brain-Computer Interfaces (BCI25)​

Unlock the potential of brain-computer interfaces through the application of neuromorphic principles, innovative signal processing algorithms for real-time decoding, and cutting-edge hardware development, enabling faster and more accurate communication between the human brain and machines.

Topic leaders - Arindam Basu (City University of Hong Kong), Pao-Sheng Vincent Sun (City University of Hong Kong), Sridhar Devarajan (Indian Institute of Science)
Read More

Organizers​

Andre van Schaik

Director ICNS, Western Sydney University, Australia

Shantanu Chakrabartty

Washington University St. Louis, USA

Piotr Dudek

University of Manchester, UK

Chetan Singh Thakur

Indian Institute of Science, India

Arindam Basu

City University of Hong Kong

Bert Shi

The Hong Kong University of Science and Technology

Steering Committee​

Our Steering Committee consists of renowned experts in neuromorphic engineering who provide guidance and strategic direction for the workshop.

Tobi Delbruck

ETH Zurich, Switzerland

Shih-Chii Liu

ETH Zurich, Switzerland

Gert Cauwenberghs

University of California, San Diego, USA

Andreas Andreou

Johns Hopkins University, USA

Giacomo Indiveri

ETH Zurich, Switzerland

Ralph Etienne-Cummings

Johns Hopkins University, USA

Jennifer Hasler

Georgia Tech University, USA

Kwabena Boahen

Stanford University, USA

Shihab Shamma

University of Maryland, USA

Rajit Manohar

Yale University, New Haven, CT, USA

Barbara Shinn-Cunningham

Carnegie Mellon University, Pittsburgh, USA

Workshop Structure

Immerse yourself in a setting that is as rigorous as it is rewarding, designed to foster innovation and collaboration among participants. Engage with experts and peers to tackle the pressing challenges in neuromorphic engineering.

Interactive Sessions and Talks​

Experience a series of enlightening talks on topics such as neuromorphic sensing, machine learning, neuroscience and chip design, delivered by leading academicians and industry professionals.

Collaborative Project Work​

Work alongside peers in team meetings, scheduled each afternoon, to brainstorm and develop project ideas. Utilize advanced research facilities to shape and refine your innovative solutions.

Pre-Workshop Preparation​

Join the dedicated Slack channel a month before the workshop starts to discuss ideas and form teams. This early engagement is crucial for a head start on project collaboration.

Key Presentation Milestones​

Prepare to present your project progress during an initial review on the 3rd or 4th day of the workshop. Culminate your experience with a final presentation on January 15th, showcasing your team's achievements to the workshop attendees.

Meet Our Speakers

We are delighted to feature an exceptional lineup of domain experts who will share their invaluable insights and expertise with us. Each speaker is a leader in their field, ready to offer a wealth of knowledge at the workshop.
Chiara Bartolozzi
Istituto Italiano di Tecnologia, Italy
Chandra Sekhar Seelamantula
Indian Institute of Science
Dhireesha Kudithipudi
University of Texas at San Antonio, USA
Chiranjib Bhattacharyya
Indian Institute of Science
Johannes Schemmel
Heidelberg University, Germany
Christian Georg Mayr
Technische Universität Dresden, Germany
Rishikesh Narayanan
Indian Institute of Science
Ernst Niebur
Johns Hopkins University, USA
Elisa Donati
University of Zurich
S P Arun
Indian Institute of Science
Supratim Ray
Indian Institute of Science
Tobi Delbruck
ETH Zurich, Switzerland
Shih Chi Liu
ETH Zurich, Switzerland
Sridhar Devarajan
Indian Institute of Science
Jenifer Hasler
Georgia Tech University, USA
Rajit Manohar
Yale University, New Haven, CT, USA
Sriram Ganapathy
Indian Institute of Science
Arindam Basu
City University of Hong Kong
Andre van Schaik
Western Sydney University, Australia
Nirupam Roy
Indian Institute of Science
Shantanu Chakrabartty
Washington University St. Louis, USA
Piotr Dudek
University of Manchester, UK
Shishir Kolathaya
Indian Institute of Science
Andreas Andreou
Johns Hopkins University, USA
Bernabé Linares-Barranco
Instituto de Microelectrónica de Sevilla
Gregory Cohen
Western Sydney University, Australia

Workshop Schedule and Session Outline

The website will soon feature a comprehensive and detailed schedule of activities. Each day will begin with a series of captivating lectures in the morning, setting the stage for an enriching experience. Afternoons will be filled with interactive work group meetings where participants will address practical challenges and brainstorm innovative solutions. Following these sessions, hands-on tutorials will enhance practical skills and deepen theoretical knowledge. The day will conclude with in-depth discussions, providing a platform to explore concepts further and address any questions.

Registration Guidelines

Registration is required to attend all events and sessions. Follow the guidelines below to secure your spot.
Application Period: Aug 1 - Sept 15, 2024
Notification of Acceptance: By Oct 1, 2024
Eligibility: Open to senior Masters, PhD students, postdocs, and industrial researchers.
Application Process: Complete the registration form. Applicants must upload a CV (maximum of two pages) including academic achievements, CGPA, and publications, and an Expression of Interest (limited to one page) outlining their current research areas, the themes in the workshop they are interested in, any specific problem statement they wish to address during the workshop, and details of two references.
Selection Process: The selection committee will create a preliminary list of applicants and send emails to the references of students for uploading letters of recommendation within 15 days of application submission. The final list of accepted and waitlisted candidates will be announced by October 1, 2024, after receiving the letters of recommendation (LoRs).
Registration Fee:
  • Students: $200 (₹16,800 INR)
  • Industrial Participants / Faculty: $500 (₹42,000 INR)
Included: accommodation and meals.
Note: The selection process described above is only for students. This process does not apply to industrial researchers, who are only required to fill out the Google form and upload their CV and Expression of Interest

 
  • Like
  • Fire
  • Love
Reactions: 9 users
Top Bottom