Customized Computing for Understanding the Brain

Project status: 

Moore’s law has been driving the exponential growth of information technology for more than 50 years, during which the ever-increased computing power has had a huge impact on people’s lives. Nowadays computers can beat humans in some specific tasks, however, they still lack behind human brains in many aspects such as lifelong learning capability and information processing efficiency, and our understanding of the brain is still at an early stage. Recently, state-of-the-art microscopy technologies provide ways to observe biological neural network at unprecedented spatial and temporal resolution, and open up new research opportunities for explaining and understanding how brain works. But with opportunities also comes challenges: Decoding neuron activity in real time with miniaturized device to support closed-loop behavioral experiments; performing neuron reconstruction, spiking neural network modeling and simulation in very large scale. It is believed that the rapid improvements in processing we have expected in the past decades to come through innovations in computer architecture, and the novel computing architecture can benefit from the knowledge gained in understanding the brain.