All the Questions About Artificial Neural Networks You Were Too Afraid to Ask

By Kelsey Campbell-Dollaghan on at

This year has seen artificial neural networks hit the big time, with Google’s Deep Dream program introducing millions of people to the concept for the very first time. But if you’re still not completely clear on, uh, how the hell they even work–fear not.

Natalie Hammel and Lorraine Yurshansky are two Google employees who have absolutely nothing to do with with deep learning–the duo work at Google’s Creative Lab, which must be about as far away from the machine learning research department as possible.

That makes them the perfect people to explain the fine-grained computer science going on in the labs of their colleagues in a new video, in which they interview people like Christopher Olah and Greg Corrado, Google researchers who work on the forefront of artificial intelligence and machine learning.

In their video, the creatives ask the scientists to explain their complex work teaching artificial brains to “see” the world. It may seem simplistic at times, but by the second half of the video, I was glad they started with the basics. If you’re interested in neural networks but don’t have the computer science background to know how they work, you’ll enjoy. And if you’ve got lingering questions, the’ll be doing an AMA on Reddit with the researchers this Friday.

[Google Research Blog h/t Prosthetic Knowledge]