Wasserstein Measure Coresets

Understand

The proliferation of large data sets and Bayesian inference techniques motivates demand for better data sparsification.

• Coresets provide a principled way of summarizing a large dataset via a smaller one that is guaranteed to match the performance of the full data set on specific problems.
• Classical coresets, however, neglect the underlying data distribution, which is often continuous.
• We address this oversight by introducing Wasserstein measure coresets, an extension of coresets which by definition takes into account generalization.