Stochastic Neighbor Embedding under f-divergences

Understand

The t-distributed Stochastic Neighbor Embedding (t-SNE) is a powerful and popular method for visualizing high-dimensional data.

• It minimizes the Kullback-Leibler (KL) divergence between the original and embedded data distributions.
• In this work, we propose extending this method to other f-divergences.
• We analytically and empirically evaluate the types of latent structure-manifold, cluster, and hierarchical-that are well-captured using both the original KL-divergence as well as the proposed f-divergence generalization, and find that different divergences perform better for different types of structure.