Dimensionality Reduction is the process of reducing the number of features in a dataset. The process is divided into feature selection (finding a subset of features of the original set which models the dataset well) and feature extraction (reducing the data to a lower dimension). Popular algorithms are: PCA, LDA and GDA. In the image below, in the left image we have a 3-dimensional image which is reduced to two 2-dimensional images.
Next: One-Shot Learning