Wasserstein Measure Coresets
arXiv:1805.07412
Abstract
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. Our formulation of the problem, which essentially consists in minimizing the Wasserstein distance, is solvable via stochastic gradient descent. This yields an algorithm which simply requires sample access to the data distribution and is able to handle large data streams in an online manner. We validate our construction for inference and clustering.
References in corpus (14)
- Auto-Encoding Variational Bayes
- Fast Computation of Wasserstein Barycenters
- A Kernel Method for the Two-Sample Problem
- Fifty Years of Pulsar Candidate Selection: From simple filters to a new principled real-time classification approach
- Sample Complexity of Sinkhorn divergences
- Coresets for Scalable Bayesian Logistic Regression
- Sliced Wasserstein Kernel for Persistence Diagrams
- Automated Scalable Bayesian Inference via Hilbert Coresets
- Stochastic Wasserstein Barycenters
- Learning Probability Measures with respect to Optimal Transport Metrics
- Parallel Streaming Wasserstein Barycenters
- One-Shot Coresets: The Case of k-Clustering
- k-Means for Streaming and Distributed Big Sparse Data
- Training Support Vector Machines using Coresets