Artificial Intelligence (AI)

Mention AI and most people will instantly think of machine learning (ML) and neural networks (NN).

There is however much more to the AI arena than these two areas. Indeed, sometimes all you might need to solve your AI problem is linear regression.

Here you will find a non-exhaustive list of some key areas where AI techniques can help us solve problems. When time allows I will attempt to create sub-pages that take a deeper dive (but still only really scratching the surface) of how AI can be implemented to solve problems in these areas.

Search

Search is concerned with how to get your AI to search for solutions to some problem... any problem. This could be simple problems to more complex problems like searching for a route in order to provide driving directions.

Knowledge

Knowledge in AI is about having your AI "know" information. It can then represent that knowledge, and use that knowledge to infer things.

Uncertainty

What happens if our AI isn't sure about a fact? i.e. we can't state for sure that a fact is true, but can we be sure to a certain probability about a fact?

Optimisation

When there's multiple ways to solve a problem, optimisation techniques help our AI find a better way, or maybe even the "best" way.

Learning

This concerns the well known machine learning (ML) arena, and learning in general by learning from data and experience.

Neural Networks

Builds on from ML and is the inspiration for AI from human intelligence.

Language

Human language processing and comprehension. This is the area of AI that includes things like being able to talk to devices and they (attempt) to understand you (speach recognition). It also involves things like natural language processing (NLP), summarisation, information extraction, language identification, machine translation, text classification and more.