A new research has revealed about a large-scale computational system, called ‘Robo Brain’, which will help robots to draw information when they need it from the internet.
Ashutosh Saxena, assistant professor of computer science at Cornell University, said their laptops and cell phones had access to all the information they wanted and if a robot encountered a situation it had not seen before it can query ‘Robo Brain’ in the cloud.
Robotics researchers further said that Robo Brain would process images to pick out the objects in them, and by connecting images and video with text, it would learn to recognize objects and how they were used, along with human language and behavior.
The system employed “structured deep learning,” where information is stored in many levels of abstraction, for example, an easy chair is a member of the class of chairs, and going up another level, chairs are furniture, but a Robo Brain knows that chairs are something you can sit on, but that a human can also sit on a stool, a bench or the lawn.
A robot’s computer brain stores what it has learned in a form mathematicians call a Markov model, which could be represented graphically as a set of points connected by lines (formally called nodes and edges).
