MNIST Exploratory Analysis
Patrick Dewey
Aanish Pradhan
Note: This page can take some time to load as it includes some computationally intensive 3D plots.
Exploratory Data Analysis
Principal Components Analysis
It is evident that linear projections only capture some of the structure of the images and that we need to extend our dimensionality reduction efforts. To accomplish this, we will leverage the \(t\)-distributed Stochastic Neighbor Embedding (\(t\)-SNE) algorithm.
\(t\)-distributed Stochastic Neighbor Embedding
