Building the Silicon Brain

Comments Off on Building the Silicon Brain
Building the Silicon Brain

As powerful as computers are, they are no match for the power of the brain. Thanks to the rapid transmission of data between the brain's 100 billion neurons and 100 trillion connections, the brain can process more information, at faster speed, than any device made of silicon and software.

In fact, even the relatively simple brain of a humble mouse is a much more efficient “computer” than anything Bill Gates or Steve Jobs ever designed. A mouse cortex works 9,000 times faster than a simulation of its functions on a personal computer—and it uses 40,000 times less energy than a PC.

But now, scientists at several institutions are making compelling progress toward artificial brains that, while still primitive, hint at the massive potential that will be unleashed once computers can be programmed to mimic the human brain.

This relatively new science is called neuromorphic engineering.1 Its objective is to use innovative technologies to create computer systems that replicate the neurons and synapses of the brain, and it is so important that governments and research institutions have made it a priority.

For example, the European Union has committed $1.3 billion to the Human Brain Project, a decade-long collaboration with the goal of simulating an entire human brain on a supercomputer.2

Meanwhile, in the U.S., a $100 million research initiative called BRAIN, which stands for Brain Research through Advancing Innovative Neurotechnologies, is focusing on developing tools to record signals from neurons in order to improve our understanding of how the human brain records, processes, stores, and retrieves information.

The success of those broad, government-funded projects will depend on the outcome of many smaller, highly focused research efforts. Consider three examples:

  1. In a project funded by the National Institutes of Health, a team of Stanford University scientists led by bioengineering professor Kwabena Boahen has developed an iPad-sized circuit board called the Neurogrid. It consists of 16 Neurocore chips that, working together, can simulate the operation of 1 million neurons, along with billions of synaptic connections. In order to reduce the energy drain that is typically involved in previous attempts at computer simulations of the brain, some of the synapses share hardware circuits on the Neurogrid. Through this innovative design, the Neurogrid uses 40,000 times less power to operate than PC-based simulations of the brain, at a speed that is orders of magnitude faster, according to an article by Boahen in the Proceedings of the IEEE...

    To continue reading, become a paid subscriber for full access.
    Already a Trends Magazine subscriber? Login for full access now.

Subscribe for as low as $195/year

  • Get 12 months of Trends that will impact your business and your life
  • Gain access to the entire Trends Research Library
  • Optional Trends monthly CDs in addition to your On-Line access
  • Receive our exclusive "Trends Investor Forecast 2015" as a free online gift
  • If you do not like what you see, you can cancel anytime and receive a 100% full refund