# Understanding Private Evolution as Learning-Augmented Clustering

> Source: <https://machinelearning.apple.com/research/private-evolution-clustering>
> Published: 2026-11-10 00:00:00+00:00

**Key takeaway:** Recasting Private Evolution as generative model-augmented Wasserstein learning makes sample complexity depend on intrinsic rather than ambient dimension, and a new geometry-aware PE variant adds provable convergence on well-clustered instances while improving recall over standard baselines.

[content type paper](https://machinelearning.apple.com/research/)published November 2026

Understanding Private Evolution as Learning-Augmented Clustering

AuthorsAudra McMillan, Kunal Talwar, Felix Zhou†**

Private Evolution (PE) is a differentially private algorithm for synthetic data generation. While it can be viewed as a Wasserstein learning algorithm, it performs much better in practice than worst-case Wasserstein analyses would predict. We recast PE as generative model-augmented Wasserstein learning. We show theoretically that when we take into account the use of a generative model that is able to capture something about the true distribution, then we can obtain much better performance bounds. For example, if the generator gives samples in the same low-dimensional space as the distribution, then sample complexity depends on intrinsic, not ambient, dimension. We also show that standard variants of PE can fail to converge on simple well-clustered instances, and propose a new geometry-aware version of PE with provable convergence on such instances. Experimentally, we show that our new algorithm is competitive with standard baselines and can improve recall.

Instance-Optimal Private Density Estimation in the Wasserstein Distance

November 21, 2024[research area Privacy](https://machinelearning.apple.com/research/?domain=Privacy)[conference NeurIPS](https://machinelearning.apple.com/research/?event=NeurIPS)

Estimating the density of a distribution from samples is a fundamental problem in statistics. In many practical settings, the Wasserstein distance is an appropriate error metric for density estimation. For example, when estimating population densities in a geographic region, a small Wasserstein distance means that the estimate is able to capture roughly where the population mass is. In this work we study differentially private density estimation…

Instance Optimal Private Density Estimation in the Wasserstein Distance

July 23, 2024[research area Privacy](https://machinelearning.apple.com/research/?domain=Privacy)[conference TPDP](https://machinelearning.apple.com/research/?event=TPDP)
