Artificial Intelligence (AI) and Machine Learning (ML) are trending fields of computer science focused on implementing intelligent systems that mimic human learning mechanisms. Many people tend to talk about AI and ML interchangeably. Although these two are correlated, they’re not the same thing and they have a lot of differences. AI is actually the bigger concept and ML is a subset of it.
Simply put, with artificial intelligence we can implement systems that simulate human intelligence. AI systems aren’t needed to be pre-programmed as these systems are able to use their own “thinking” instead. Good examples of AI applications e.g. Face Recognition and Text Correction algorithms.
Artificial Intelligence can be split into 3 different categories based on what their capabilities are:
- Weak AI
- General AI
- Strong AI
At the time of writing, most of the AI-based systems are using weak AI and general AI. In the future, strong AI-based systems are going to be able to think and carry out tasks on their own, just like us humans do.
ML systems are able to learn from past data and experiences and no explicit programming is needed. ML-based computer systems make predictions based on historical data. These systems are utilizing different types of Machine Learning models that use vast amounts of structured or semi-structured data in order to make accurate predictions based on it.
As a simple example, we could make an ML-model predict whether there’s an apple or orange in an image by “showing” a lot of images of both and making the model extract useful characteristics of the images, such as color.
Three main categories of ML involve:
- Supervised learning
- Unsupervised learning
- Reinforcement learning