The field of artificial intelligence can be daunting to outsiders. News and pop culture have created the illusion that the field is a ticking timebomb for a dystopian explosion of superintelligence, while in reality that almost certainly is not the case. The A.I. systems that can detect faces cannot also understand speech, and that simple fact means that even the most advanced artificial intelligence systems won’t take over the world in the near future.
One of the reasons why people can be scared of A.I. is because it’s difficult to understand. Facebook’s director of artificial intelligence, Yann LeCun, uses the analogy that A.I. is a black box with a million knobs; the inner workings are a mystery to most. But now, we have a peek inside.
Masters candidate at Ryerson University Adam Harley has built an interactive visualization that helps explain how a convolutional neural net, or artificial intelligence program best used in figuring out images, works internally.
Artificial neural networks are the flavor du jour of A.I. right now, and are mostly what big companies like Facebook and Google use when they talk about employing artificial intelligence. Convolutional neural nets, as depicted in Harley’s interactive, were invented by Facebook’s Yann LeCun back when he worked at Bell Labs in the 1980’s. They’re just one type of artificial neural network.
As seen in the interactive, neural networks work in layers, or differing levels of abstraction. There’s the input layer, the idea the computer is trying to make sense of, and the output layer, the computer’s final conclusion. In between are layers of mathematical functions, each layer condensing the most important distinguishing information and passing it to the next layer, which focuses on other details. At the end, based on the setup of the network, the computer will be able to formulate an answer. In Harley’s model, the computer can simply distinguish a number, much like the original convolutional neural nets used to read check deposits in ATMs. Cutting edge AI is far more complex, able to recognize faces with 97 percent accuracy.
But seeing is believing, try out artificial intelligence for yourself!
[H/t Samim Winiger]