Principal Investigators: Bilge Yildiz, Michale Fee, Jesus del Alamo, Ju Li

Neuroscience-guided ionic computing: The brain is capable of information processing on a massively parallel scale with energy consumption of 1 – 100 fJ per synaptic event. New approaches to brain-inspired computing could present opportunities to achieve greater than a million-fold improvements in energy efficiency. The goal is to translate the understanding of learning rules in the brain, to the design of brain-guided, energy-efficient platforms. The work should design, implement, and test novel hardware architectures, devices, and materials that emulate the neural circuits and synaptic plasticity rules in learning behaviors, which require sensing, reasoning, and action. At the device level, inspiration is the biological synapse, which is an ultra-efficient electrochemical machine working with ions in liquid medium, while combining processing and memory in one unit. Given that ions can also be modulated electrochemically in solid state (as we readily do in batteries and fuel cells), a promising direction for the field is to establish the ability to process information with ions in solid state, including the ions involved in neurotransmission. The Ionic computing approach has the potential to compete with, and even surpass, the energy efficiency of the brain. Computing with the neurotransmission ions may also pave the way to interfacing such hardware with the brain itself.