5. Dimensionality

5.1. Pre-Reading

• The Curse of Dimensionality: Why High Dimension Data Can be So Troublesome

I also need you to understand what the mathematical term tensor is.

A tensor is a generalization of scalars, vectors, and matrices to an arbitrary number of indices.

This guy is very good but you may want to speed up the video a tad:

• If you are new to linear algebra, watch the entire video What’s a Tensor

• If you have experience in linear algebra, start at 6:15

• If you are seriously in a hurry watch at least from 10:20

5.1.1. Objectives

1. Features exist in dimensions

2. More dimensions have consequences (data and compute requirements)

3. There exists ways to reduce dimensions