The rising demand for faster, more energy-efficient computation presents a rare opportunity to develop novel architectures and hardware using alternative materials. Out of this need, the field of neuromorphic computing has emerged with inspiration taken from the parallelism and robustness of the human brain. Of the existing neuromorphic architectures, spiking neural networks are perhaps the most bio-realistic approach, mimicking the spiking dynamics of neurons to attain superior energy efficiency with the additional benefit of providing temporal information.
In this talk, Dr. Toomey presents a power-efficient artificial neuron made from superconducting nanowires, which naturally generates spiking based on the nonlinear transition between the superconducting and resistive states. The device experimentally reproduces multiple bio-realistic behaviors, while simulations are used to explore how networks of nanowire neurons may be used in inference or for studying theories of how biological neurons interact.
Bio: Dr. Emily Toomey currently works as a researcher at Station Q as part of Microsoft's quantum team. Prior to joining Microsoft, she worked at the Laboratory for Physical Sciences (LPS) where she studied hybrid superconductor-semiconductor devices. Emily completed her PhD in Electrical Engineering at MIT as an NSF Graduate Research Fellow, where she developed memory cells and artificial neurons using superconducting nanowires (the topic of this talk). At the heart of all of her research is a passion for understanding and exploiting unique dynamics in nanoscale material systems in order to create new devices for energy-efficient computing. Outside of research, Emily is involved in science communication and journalism, and spends most of her time oil painting.