# Predictive Coresets

arXiv: https://arxiv.org/abs/2502.05725 · HTML: https://mlsanity.com/paper/2502.05725/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 32 references that are in the map.
- First posted on arXiv 8 Feb 2025.

## Builds on (most cited first, 11 of 11)

- [Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm](https://mlsanity.com/paper/1608.04471/) (2016)
- [Coresets for Scalable Bayesian Logistic Regression](https://mlsanity.com/paper/1605.06423/) (2016)
- [Automated Scalable Bayesian Inference via Hilbert Coresets](https://mlsanity.com/paper/1710.05053/) (2017)
- [Streaming Variational Bayes](https://mlsanity.com/paper/1307.6769/) (2013)
- [Martingale posterior distributions](https://mlsanity.com/paper/2103.15671/) (2021)
- [Approximate Bayesian computation with the Wasserstein distance](https://mlsanity.com/paper/1905.03747/) (2019)
- [Nonparametric learning from Bayesian models with randomized objective functions](https://mlsanity.com/paper/1806.11544/) (2018)
- [Concentration bounds for linear Monge mapping estimation and optimal transport domain adaptation](https://mlsanity.com/paper/1905.10155/) (2019)
- [Subsampling MCMC - An introduction for the survey statistician](https://mlsanity.com/paper/1807.08409/) (2018)
- [Machine Learning and the Future of Bayesian Computation](https://mlsanity.com/paper/2304.11251/) (2023)
- [Bayesian predictive inference without a prior](https://mlsanity.com/paper/2104.11643/) (2021)

## Abstract (arXiv)

Modern data analysis often involves massive datasets with hundreds of thousands of observations, making traditional inference algorithms computationally prohibitive. Coresets are selection methods designed to choose a smaller subset of observations while maintaining similar learning performance. Conventional coreset approaches determine these weights by minimizing the Kullback-Leibler (KL) divergence between the likelihood functions of the full and weighted datasets; as a result, this makes them ill-posed for nonparametric models, where the likelihood is often intractable. We propose an alternative variational method which employs randomized posteriors and finds weights to match the unknown posterior predictive distributions conditioned on the full and reduced datasets. Our approach provides a general algorithm based on predictive recursions suitable for nonparametric priors. We evaluate the performance of the proposed coreset construction on diverse problems, including random partitions and density estimation.