Artificial Intelligence (AI) is created to emulate the human brain, the most powerful and intelligent natural computer known to us. Neuromorphic computing is a special technological branch that explores ways to build intelligent machines that emulate human and other mammal brains.
At the Indian Institute of Technology, Delhi (IIT-Delhi), one of India’s premier-most engineering colleges, has developed a new spiking neuron model named DEXAT (Double EXponential Adaptive Threshold). This invention, led by Prof Manan Suri, Department of Electrical Engineering, has opened the doors to possibilities of creating more accurate, faster and more energy-efficient neuromorphic artificial intelligence (AI) systems for real-world applications like speech recognition.
“We have successfully demonstrated the utilisation of memory technology beyond simple storage. We have efficiently utilised semiconductor memory for applications such as in-memory computing, neuromorphic computing, edge-AI, sensing and hardware security. This work specifically exploits analogue properties of nanoscale oxide-based memory devices for building adaptive spiking neurons,” Suri says in an IIT-Delhi press release.
As observed in studies, DEXAT has exhibited higher accuracy, faster convergence and flexibility in hardware implementation compared to other state-of-the-art adaptive threshold spiking neurons. In addition, the model is able to achieve high performance with fewer neurons, making it energy efficient.
The scientists successfully demonstrated a hybrid nanodevice-based hardware realisation. The proposed nanodevice neuromorphic network was found to achieve 94 per cent accuracy even with very high device variability, indicating robustness.
Source: indiaai.gov.in
