[Hot] 2016 hot words: machine learning and artificial intelligence can you distinguish?

According to the American Wired magazine website, thanks to the strong promotion of Silicon Valley giants such as Google, Amazon and Facebook, artificial intelligence (AI) and machine learning have never been known to the public. In the concept of the average person, these two fashionable words seem to be used interchangeably, referring to advanced productivity that can replace human work.

However, although AI and machine learning are inextricably linked, the two are not the same thing.

AI is a branch of computer science, known as Artificial Intelligence, which aims to build intelligent machines by hand. Stanford University defines "machine learning" as the ability to make a computer operate autonomously outside of manual programming. The two are inextricably linked, you need AI experts to build smart machines, and the participation of machine learning experts is indispensable for real intelligence.

Both Google and Nvidia are on the road to exploring machine learning. Their vision is to allow machines to eventually learn and think like humans, which is considered the key to the next wave of technological revolution.

The public is no stranger to the results of machine learning. In the past few decades, machine learning has spawned the fruits of “self-driving cars”, “precise speech recognition” and intelligent web search for the benefit of all. It even helps humans crack genetic code. However, under this series of incredible results, how does machine learning work?

Let's take a web search as an example. Let's say you want to search for "Wired" in Google, but you accidentally made it "Wored." In the case of a non-intelligent engine, the result of the return will of course make you aware of your mistakes, and then you will spend a few more seconds to search again with the correct spelling. Google's algorithm will record your "secondary search" situation. When someone makes the same spelling mistake later, Google will ask "Do you want 'Wired'?" and give the correct search. result. So, Google learned the automatic correction function.

The above is just a very basic "machine learning" case. In reality, scientists and researchers use more sophisticated machine learning to achieve things that were previously impossible. Programs learned from human experience can help scientists uncover the mystery of the operation of genes. Machine learning is used commercially to provide more accurate advertising and buying advice through analysis of consumer behavior, thereby greatly increasing business profits. Other applications include image recognition, fraud prevention, and more.