What is 'T-SNE'?

What is 'T-SNE'?

Approach: Describe a non-linear dimensionality reduction technique for visualization.

t-Distributed Stochastic Neighbor Embedding. A non-linear dimensionality reduction technique well-suited for embedding high-dimensional data for visualization in 2D or 3D space.

Follow-up: Why is it better than PCA for visualization?

Key concepts: manifold learning, kullback-leibler, probability distributions.