SyNAPSE: Jun Sawada

Since its invention, the digital computer improved significantly in performance and storage capacity. However, it remains to be machinery-although very complex one- that has to be programmed in every detail by programmers. Even though parallel systems have become popular, majorities of processor cores remain sequential in nature. Comparatively, the human brain can perform complex tasks rapidly and accurately using energy as little as 20 watts without being programmed. And it is fundamentally a parallel computer. Making sense of real-time input flowing in at a dizzying rate is a Herculean task for today´s computers, but would be natural for a brain-inspired system. By reproducing the structure and architecture of the brain-the way its elements receive sensory input, connect to each other, adapt these connections, and transmit motor output-the cognitive computing systems could potentially emulate the brain´s computing efficiency, size and power.