Wiki Categories

Model Evaluation

Supervised Learning

Supervised learning, in the context of artificial intelligence (AI) and machine learning, is a system that provides both input and output data. Input and output data are then tagged for classification to establish a learning base for further data processing.

Why use supervised learning ?

The data used for supervised learning is a series of examples including pairs of input subjects and expected outputs (also known as supervisory signals). Take the example of a supervised learning system for image processing, in which photos of vehicles belonging to the car and truck categories are introduced. After sufficient observation time, the system must be able to distinguish between multiple unlabeled images and categorize them. Once this goal is reached, the learning can be considered complete.

Supervised learning models have some advantages over unsupervised models, but they also have some limitations. For example, they are more likely to make relatable decisions because they rely on data provided by humans. However, supervised learning systems have difficulty processing new information as recovery-based solutions. If a system that knows two categories - cars and trucks, and it receives a bike image, it will have to place it in one of these two categories, which will be incorrect. But if the system is generative, it will still not necessarily recognize a bike, but it can identify it as a different category.


Additional Resources:

Explorium delivers the end-game of every data science process - from raw, disconnected data to game-changing insights, features, and predictive models. Better than any human can.
Request a demo
We're Hiring! Join our global family of passionate and talented professionals as we define the future of data science. Learn More