Purdue University created a new research initiative called the Center for Brain-inspired Computing Enabling Autonomous Intelligence, or C-BRIC with the mission to deliver key advances in cognitive computing, with the goal of enabling a new generation of autonomous intelligent systems such as self-flying drones and interactive personal robots.
The center intends to develop the next generation of neuro-inspired algorithms and theory, neuromorphic computing fabrics and distributed intelligence.
Their vision is to enable next generation of intelligent autonomous systems by harnessing the key information processing principles of the brain.
Under Neuro-inspired algorithms, they intend to develop bio-plausible network architectures and learning algorithms for perception, reasoning, and decision making.
If only brain reduced to algorithms…
If you were to design a brain, where would you start?
To make a decision, brain must have a model of the world and a model of its own body relative to the world. It must filter in (not out) only the relevant and important parameters from the world and from the body that will interact based on what the brain predicts. The brain must expect how the world will change at several milliseconds interval rate in order to prepare its prediction and response. And then it has to have finest control of the body (think about your fingers) to manipulate the world. And all of that must be remembered including times and places where it happened. And the energy budget for action is under 80 watt. And here’s a little secret: the model of the world is the world, it is not a copy of the world in the brain; it is out there and not inside of the brain!
Can you write an algorithm for this? What would you reduce the problem to in order to make it manageable?
I wrote about unsolved puzzles of consciousness. The great news is that we can ask these questions!