BRN Discussion Ongoing
- Thread starter TechGirl
- Start date
Frangipani
Top 20
#neuromorphic #iot2025 #aiot #neuromorphiccomputing #edgeai #sustainability #iiot #energyefficiency #research | Behrooz Azadi
Next week, I’ll be presenting our paper “Towards an Energy-Efficient and Sustainable IIoT using Embedded Neuromorphic AI” at The 15th International Conference on the Internet of Things (IoT 2025) in Vienna. Our work explores how #neuromorphic hardware can make Industrial IoT more...www.linkedin.com
View attachment 92893
View attachment 92901
19 November 2025 - Session 2:
“Towards an Energy-Efficient and Sustainable IIoT using Embedded Neuromorphic AI
Behrooz Azadi, Bernhard Anzengruber-Tanase, Georgios Sopidis, Michael Haslgrübler and Alois Ferscha”
View attachment 92894 View attachment 92895
View attachment 92896
About Us – Pro²Future
www.pro2future.at
View attachment 92897 View attachment 92898
Shareholder – Pro²Future
View attachment 92899
What we offer – Pro²Future
View attachment 92900
I recall somebody posting the below Pro²Future poster the other day, some of whose co-authors are also co-authors of the above paper that is going to be presented at the conference in Vienna next week:
View attachment 92903
The above-mentioned IoT25 conference paper “Towards an Energy-Efficient and Sustainable IIoT using Embedded Neuromorphic AI” by Behrooz Azadi, Bernhard Anzengruber-Tanase, Georgios Sopidis, Michael Haslgrübler (all Pro2Future GmbH, Linz) and Alois Ferscha (Johannes Kepler University Linz) can now be accessed online:
“Neuromorphic hardware offers a promising alternative, with the potential for much lower energy consumption and inference latency compared to GPU-based com- puting. Throughout this study, we investigate this claim and compare the energy consumption and inference latency of BrainChip Akida and NVIDIA Orin NX. Our results indicate that the quantized model runs faster on the Akida than on the Orin for inference, with an average of 22.54 seconds compared to 181.66 seconds for 10k test samples from the MNIST dataset, respectively. Furthermore, the measured active power and, therefore, energy consumption associated with Akida during idle time are considerably lower than that of NVIDIA Orin. Additionally, our uncontrolled long-term monitoring confirms that the neuromorphic hardware consumes significantly lower energy over 25 days, with an average active power of 4.36 W compared to 9.17 W and energy consumption of 2.6 kWh compared to 5.5 kWh.”
[…]
[…]
Similar threads
