(GIST OF SCIENCE REPORTER) IIT Delhi Researchers Demonstrate a New Brain-inspired Artificial Neuron for Building Accurate and Efficient Neuromorphic AI Systems
IIT Delhi Researchers Demonstrate a New Brain-inspired Artificial Neuron for Building Accurate and Efficient Neuromorphic AI Systems
- Researchers at IIT Delhi have invented a new spiking neuron model named as DEXAT (Double EXponential Adaptive Threshold neuron).
- The invention is significant as it will help to build accurate, fast and energy-efficient neuromorphic Artificial Intelligence (AI) systems for real-world applications like speech recognition.
- The work being interdisciplinary lies at the intersection of Artificial Intelligence, Neuromorphic Hardware and Nanoelectronics.
- The study demonstrated neuron model with higher accuracy, faster convergence and flexibility in hardware implementation compared to other state-of-the-art adaptive threshold spiking neurons.
- The proposed solution achieves high performance with fewer neurons. Benefits of proposed invention were shown on multiple datasets.
- Classification accuracy of 91% on Google Spoken Commands (GSC) dataset was achieved. A patent has also been filed on this work.
- The proposed nanodevice neuromorphic network was found to achieve 94% accuracy even with very high device variability indicating robustness.
Courtesy: Science Reporter