A research team led by Prof. Shahar Kvatinsky has developed innovative hardware that accelerates the learning process of artificial intelligence systems.

Dr. Kvatinsky’s research group has developed, on a theoretical level, hardware systems specifically designed to work with these networks,

allowing the neuron network to perform the learning phase at high speed and with reduced energy.

“Compared to working with graphics processors, our hardware improves the speed of computation by 1,000 and reduces energy consumption by 80 percent,” Kvatinsky.

Academic Article

Photo: Nitzan Zohar