Every conversation these days is littered with words like “artificial intelligence,” “machine learning,” and “deep learning.” It’s like failing to catch up with the Internet in the 1990s.
But, to put it bluntly, few really know the business, really few.
A few months ago, I was one of those people who could more or less jump on the bandwagon, but didn’t really understand what artificial intelligence was all about. If anyone tells me about the technical details, I will be devastated: shut up.
However, a few videos completely changed my understanding, which used extremely easy to understand language and cartoon pictures, let me understand artificial intelligence, machine learning in the end what the ghost, particularly interesting. On that basis, I started to really grow in the field.
If you’re like me and don’t know much about machine learning but want to learn something dry without going through a lot of trouble, check out these videos I’ve shared and you might become a semi-expert in a few months.
Based on my experience, I’m going to divide these videos into three categories: the first category is for beginners, the second category is for basic, and the third category is for advanced. I promise, each category is easy, and you won’t have to work too hard.
Here, I would like to make a small note that due to the limitation of wechat, only three videos can be implanted in one article, so other videos can only be attached with links, but under the condition of wechat, this link cannot be opened directly. To make it easier for you to read, you can click on the original text to open the link.
Introduction – Basic introduction
Before recommending the starter video, let me give you a few words to make sense of the combination. (Key learning area of Xiaobai)
What is the purpose of artificial intelligence?
A: To make machines behave more like humans and even surpass them in certain skills.
To do this, AI must “think,” and its thinking includes techniques and methods for solving computing problems in an intelligent way. Sounds a little abstract. A few examples: search – find the fastest round trip between two cities, planning – let the robot navigate and achieve a given goal……
Of course, the ability of ARTIFICIAL intelligence depends on two conditions: algorithm and computing power.
How good the algorithm is (you can think of it as how good the machine is at thinking), and how good the computing power is (you can think of it as the supply of blood, protein and other logistics for the machine to think), these two conditions directly determine how good the ARTIFICIAL intelligence you build is.
As long as we humans keep the algorithms and computing power, there’s nothing else to worry about. The machine will learn by itself.
How does a machine learn for itself?
A: Solve problems by learning from existing annotation data without hard-coding rules into the algorithm.
It’s still too abstract. Let me give you an example.
For example, develop a system that can identify dogs and cats in pictures. To do this, we simply put a lot of pictures of cats and dogs into the algorithm we’ve set up and go to sleep. By the time the machine has finished looking at this mass of images, it will know what a cat is and what a dog is.
So what is deep learning?
Deep learning is machine learning, but it’s more complicated. It is similar to installing a brain in a machine, which also has a multi-layer system made up of a large number of neurons, and can handle complex machine learning tasks (such as cat and dog image recognition mentioned above) with a large amount of data training.
Why is deep learning all of a sudden so popular?
Because the two hardware and software it needs have been met: in terms of hardware, GPU and other functions have become powerful, and the price has dropped significantly; On the software side, there’s a lot more data to train models than ever before.
Well, with all that information out of the way, let’s start our introductory video tour.
Artificial Intelligence, Deep Learning, and Machine Learning: An Introduction by Andreessen Horowitz
Video address: v.qq.com/x/page/g050…
This video is about the history of artificial intelligence & Deep learning introduction
Introduction to Machine Learning (Udacity)
Video address: v.qq.com/x/page/w050…
A brief (slightly more technical) explanation of machine learning
Artificial intelligence is the new electricity (Andrew Ng)
Video address: v.qq.com/x/page/b050…
Ng explains the current state of ai/machine learning, its applications and management measures
Ng also posts lots of funny videos on Youtube, and has a free machine learning course on Coursera, which is highly technical/theoretical. You can search for children’s shoes if you need them.
Growth — Learn more
If you want to dig deeper into the technical details, you can watch the Deep Learning Simplified video series on Youtube and a Facebook video explaining machine Learning/Deep Learning.
Ai revealed (Facebook) :
Yann LeCun on artificial intelligence
Video address: v.qq.com/x/page/y050…
Yann LeCun introduces deep learning
Video address: v.qq.com/x/page/h050…
For more technical details, check out the Facebook video’s official page.
Deep learning
Deep Learning Simplified – Getting started
Video address: v.qq.com/x/page/g050…
To check out all the videos in the series, head over to its Youtube channel.
Advanced – Build your own neural network
To go ahead and learn and build your own neural network, you need to:
1. Learn how to program Python (take classes at Codecademy, Codeschool or Udacity)
2. Learn about Google’s TensorFlow framework or Keras to find a simpler model implementation method; Then follow the MNIST tutorial to get a general idea of machine learning; I highly recommend the following video.
Video address: v.qq.com/x/page/u050…
5 minutes X learns Tensorflow
3. Also, take a look at Google’s free deep learning course on Udacity. Do your own search.
This is the end of my sharing, in addition to these, if you also have good video resources, welcome to the background message recommendation oh.
The original link