Google’s DeepMind Learns How to Navigate the Tube

It’s something that some Londoners, and many tourists, struggle with: navigating the London Underground without getting lost. Now DeepMind, which Google acquired in 2014, has invented a computer that has independently learnt to do just that through deduction. DeepMind’s new artifical model is called a differentiable neural computer (DNC). It is fed information on how items and data are related to each other: for example, a map of the London Underground network or a family tree. It is given an external memory of the information. It then answers questions about the relationships within those structures. The memory of those answers help the AI computer to infer new answers, in other words to “learn”: While impressive, what applications could this have for the rest of us? To begin with, it could help artificial intelligence become more powerful, and therefore, more useful in everyday problem solving. An example in commercial use already are digital assistants like Google’s Voice and Apple’s Siri, which deduce audio patterns to enable speech recognition. Other possible applications could be to help you navigate…

Link to Full Article: Google’s DeepMind Learns How to Navigate the Tube

Pin It on Pinterest

Share This

Join Our Newsletter

Sign up to our mailing list to receive the latest news and updates about and the Informed.AI Network of AI related websites which includes Events.AI, Neurons.AI, Awards.AI, and Vocation.AI

You have Successfully Subscribed!