Skip to contentSkip to filters

We use cookies for analytics.

How Neuromorphic Computing Can Help Us Understand the Brain

James Bradley Aimone

Sandia National Laboratories

Abstract

Neural dynamics span a wide range of temporal and spatial scales: neurological and psychiatric disorders unfold over months to years, while most research concentrates on short timescales, and the computational cost of simulation limits brain models to seconds of activity. Inspired by efforts to model adult neurogenesis in the hippocampus, James Bradley Aimone turned to neuromorphic computing. Neuromorphic systems are now reaching brain-like numbers of neurons, and he describes unexpected results on both cognitive and non-cognitive tasks. The hardware could enable energy-efficient AI and real-time biomedical computing with strict demands on performance, efficiency and security, and understanding neural computation at biologically realistic scales can help explain how and why the brain works.

Topics