Artificial Intelligence is everywhere now, what changed?
This artificial intelligence tutorial explains in a simple way by comparing how a child learns how to recognize various objects & why we see AI everywhere now.
00:00 So why is artificial intelligence everywhere all of a sudden? So Hi, this is Charlotte from fashion marketing and in this tutorial, I’m going to talk about why are we seeing artificial intelligence, machine learning deep learning everywhere and what has actually changed it caused this. So to understand that, let’s take a look at how a child actually learns. So when you have a child, a niece and nephew or your own kid probably, and you want your kid to tell the difference between a car and a bus. So what do you do? You show them multiple images of a car. Like, let’s say we’re driving on the freeway and you’re like, hey, that’s a car that’s a car. And then you see a bunch of buses and you tell them, hey, that’s an image of a bus or this is actually a bus looks like. So you’re teaching them and eventually you’re showing them so many images of cars and buses and telling them this is actually what a car looks like or this is actually what a bus looks like.
00:50 And eventually the kid learns to tell the difference between a car and a bus, and now when you show them an image of a car or a bus that they have not seen before, and you ask them, hey, which one is the bus? And if they’re able to get this right, that this is a bug and this is, this is not a bus, you know, that, that per, that your child is actually learning. And now in the future, if you show them an image of a bus or a car that they have not seen, they will be able to tell the difference. So this is basically how machine learning or deep learning is used to teach computers so they can recognize any object and they can tell one object from another. So in, in order for artificial intelligence to function, it basically needs three components, right?
01:34 So number one is high-performance GPU, which can calculate a fast enough. Number two is sophisticated algorithms that you can create a sophisticated model that tells what is the definition of a bus and what is, how do you define a bus? Or if I were to define a pair of shoes or things like that. And third is it lots of labeled data. So which also has like two subcategory, which means having a lot of data and then having a lot of labeled data. So in the past few years we always had, you know, we had high-performance Gpu, Nvidia, we always can create sophisticated algorithms because the programming language has not changed that much, but what would need, what we did not have was a lot of big data and lot of labeled data. So what, what is labeled data? So you think about, you know, 90 percent of the world data has been created in last two years and two point three quintillion bytes of data is created every single day.
02:32 So what is a label data? So labeling data basically means is the way you were teaching your kid, right? So you were telling him this is a car, this is a bus. So in the same way, think about having thousands and thousands and millions of images and each image actually being labeled that this is a car, this is a person you know, this is a sign, this is a signal, and the same thing was done for clothing that this is actually a mustard solid full wool coat. Know these are sneakers and they are solid, they have laces, they are pink in color or this is a sweatshirt which has yellow in color, has drowned as full and as a solid color. So a lot of this data has been available, which has turned out to be big data and big data. That data has been labeled by either by a lot of companies or lot by people.
03:21 And the biggest example is facebook. So remember back in the days when Facebook came out and it was asking you to tag your friends, it could not tell which was the person’s face. So we will just point at where a regular person and we’ll ask you to click on the person’s face and then you typed in the name of the person that this is carried. This is kate, fast forward a couple of years. Then facebook was able to tell that this is actually a face, but it still could not tell whose face it is, so it will put a little square around the face and they’ll ask you to tag the person that you will still tie it into. This is Carrie. This is kate. This is Nicole, and now facebook has had enough people on the facebook platform. Tag carries image so many different times from so many different angles and now facebook is actually able to figure out that this actually carries and this is kate and this is Nicole and this is Jessica.
![Artificial Intelligence Can Clone Any Voice, Yours Too [ LyreBird ]](https://video.kidibot.ro/wp-content/uploads/2020/01/1544/artificial-intelligence-can-clone-any-voice-yours-too-lyrebird-150x150.jpg)