Learning methods

Algorithms

Computers follow steps(algorithms) to process input into desired output. As clever as those roadmaps sometimes are, they still can't match human intelligence. Moreover, all those algorithms were first devised by humans. In fact, computers do not yet succeed in thinking through all the steps themselves or in arriving at the best possible solution to problems. Humans can. Even though you make quite a few wrong choices in your life, in many cases you learn from your mistakes. Of course, we don't learn everything by making mistakes first. We learn most knowledge and skills through experience or by looking at examples. Repetition helps with this, we all know that. 

Machine learning 

Artificial intelligence means that computer systems learn by themselves to arrive at solutions. We program software to learn on its own through experience or by example, just as humans and even animals do. Computer scientists use a variety of learning algorithmsto do this. For example, they try to simulate the functioning of neurons in the human brain with software. We speak of a "neural network" in this case. Most things in our lives we too learn by examples from others. Even in school we have to learn from examples and mistakes. 

• Online/batch learning (=continuous learning....)
• Supervised learning (supervised learning)
• Unsupervised learning (unsupervised, unmonitored learning)
• Reinforcement learning
• Hybrid learning

Supervised learning

Suppose you have a pet cat and you keep telling your two-year-old son, "This is the cat. However, your little son has never seen a dog before. Chances are that upon his first encounter with a dog, he will say, "Look mommy, cat! You will then say, "No way, that's a dog. By repeating this several times, your little son will quickly learn the difference between a dog and a cat. He will gradually learn to distinguish between the two and then also understand that the neighbor's sheep are neither dogs nor cats. Children (even at school) learn by 'supervised learning,' or learning under supervision. They learn from their mistakes and from examples because someone points them out. By programming neural networks into software, computers can learn on their own ('machine learning'). 

We first feed the AI software with  training sets  (or  datasets ). For example, these could be a series of pictures of dogs. By feeding a mass of pictures of dogs as training sets to the software, the neural network gradually learns to recognize dogs in other pictures by itself. A key difference is that a human needs fewer examples than an AI algorithm. A child who has seen a cat a few times will also quickly recognize other cats.   

AI needs a mass of examples.After all, a cat can look very different, as there are dozens of breeds, for example. In addition, you also need to be able to recognize a cat from all possible "angles" (top view, back view...) and among other objects in an image. For this, an AI must also be able to segment or distinguish various objects from each other. After all, when you show an image or webcam image to an AI, that input data consists only of a series of pixels with RGB color values. So the AI has to look for patterns in pixel data.

Man-woman recognition. Check it out:  https://www.leerschool.be/tools/manwoman/

Reinforcement learning

In 'supervised learning,' the algorithm learns from examples. In reinforcement learning, you reinforce the algorithm by rewarding or punishing it. If the predicted output is wrong or incomplete or leads to the wrong result, the AI software can 'learn' from its own mistakes. Suppose a robot is driving around in a room that has only one exit, that robot may try to get out by randomly bumping into the wall until it finds the opening. But once the robot has discovered the door, then it can 'memorize' it. 

Unsupervised learning 

In unsupervised learningthe algorithm is not given examples, but has to classify (categorize) elements on its own based on common features. In unsupervised learning, the algorithm is not given examples or labels beforehand. The algorithm must classify the information itself based on features. Suppose you have to do laundry for the first time in your life. You see a pile of unwashed clothes in front of you. You now have to classify the laundry based on a number of characteristics. Classifying by color and textile type would be the desired output. But that is what the "algorithm" now needs to discover. If you use the K-means algorithm, you can only specify in advance how many groups you want. The rest the algorithm has to discover for itself.