/en/the-ai-revolution/types-of-ai/content/

Machine learning

Machine learning plays a significant role in Artificial Intelligence. It involves developing algorithms and models that allow computers to learn on their own. It’s used in many applications, from image and speech recognition to making predictions and decision-making.

Computer vision is a part of AI that helps computers understand and make sense of images and videos. These AI systems often use "supervised learning" (explained below) to recognize objects, identify different parts of an image, and perform other visual tasks effectively. In a sense, it tries to replicate human visual perception.

There are three main types of machine learning: supervised, unsupervised, and reinforcement learning.

Supervised learning: In supervised learning, an algorithm is trained on what’s called “labeled data.” This just means that the data is already associated with the correct output or answer. The AI uses labeled data to identify patterns, in turn making predictions about information it's never seen before. Examples of supervised learning include understanding images and speech recognition.

Unsupervised learning: Unsupervised learning is when a computer explores the data on its own to find hidden patterns and structures, without being told what to look for. It does this by grouping similar information, or looking for ways to break it down into something less complicated. Some examples of unsupervised learning are recommendation systems, cybersecurity, and financial forecasting.

Reinforcement learning: This is when a computer program learns by interacting with its surroundings, and getting either a positive or negative response. It tries to learn which actions will give it the most positive feedback by learning from its mistakes—sort of like trial-and-error. Reinforcement learning can be used to teach computers how to play games, or to control robots.

Continue on to the next lesson to learn about neural networks.

/en/the-ai-revolution/neural-networks/content/