Exploring Light-based Neuromorphic Computing for AI Applications with Patty Stabile

How to create neural networks inspired by the brain using light

Supercomputers can be extremely fast but they also consume a lot power. Neuromorphic computing can be a viable alternative, as it uses our brain to create fast, energy-efficient computers. This technology offers a wide range of applications, including autonomous driving, medical image interpretation, edge AI, and long-range optical communications. Patty Stabile, an electrical engineer, is a pioneer in exploring new computing paradigms that are inspired by the brain and biology. \”TU/e has all the ingredients to demonstrate photon-based neuromorphic computation for AI applications.\”

Patty Stabile is an associate professor at the Department of Electrical Engineering. She was one of the first people to explore the new field of photonic neuralmorphic computing.

I was working on a proposal for photonic digital artificial neuron when, in 2017, researchers from MIT published a paper describing their development of a small chip that could perform the same algebraic functions in an analog manner. It was then that I realized analog synapses were the best way to run artificial intelligence. I’ve been fascinated by the topic ever since.