# Weighted quantization using MMD: From mean field to mean shift via gradient flows

Source: [Weighted quantization using MMD: From mean field to mean shift via gradient flows](<https://www.sota2.com/research/paper/68fa5d791e209131ce9da66f>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-25T05:00:11.494Z

Authors: Ayoub Belhadji, Daniel Sharp, Youssef Marzouk

Published: 2025-02-14T23:13:20.000Z

## Abstract

Approximating a probability distribution using a set of particles is a fundamental problem in machine learning and statistics, with applications including clustering and quantization. Formally, we seek a weighted mixture of Dirac measures that best approximates the target distribution. While much existing work relies on the Wasserstein distance to quantify approximation errors, maximum mean discrepancy (MMD) has received comparatively less attention, especially when allowing for variable particle weights. We argue that a Wasserstein-Fisher-Rao gradient flow is well-suited for designing quantizations optimal under MMD. We show that a system of interacting particles satisfying a set of ODEs discretizes this flow. We further derive a new fixed-point algorithm called mean shift interacting particles (MSIP). We show that MSIP extends the classical mean shift algorithm, widely used for identifying modes in kernel density estimators. Moreover, we show that MSIP can be interpreted as preconditioned gradient descent and that it acts as a relaxation of Lloyd's algorithm for clustering. Our unification of gradient flows, mean shift, and MMD-optimal quantization yields algorithms that are more robust than state-of-the-art methods, as demonstrated via high-dimensional and multi-modal numerical experiments.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Quantization on Rings distribution](<https://www.sota2.com/research/sota/quantization-on-rings-distribution.md>): Quantization; Rings distribution; MMD: 0.0815.

- [Quantization on Checkers distribution](<https://www.sota2.com/research/sota/quantization-on-checkers-distribution.md>): Quantization; Checkers distribution; MMD: 0.068.

- [Quantization on Funnel 2D Distribution](<https://www.sota2.com/research/sota/quantization-on-funnel-2d-distribution.md>): Quantization; Funnel distribution Neal's two-dimensional; MMD: 0.0257.

- [Dataset Summarization on GMM-100](<https://www.sota2.com/research/sota/dataset-summarization-on-gmm-100.md>): Dataset Summarization; GMM-100; MMDw: 0.0282.

- [Dataset Summarization on Joker](<https://www.sota2.com/research/sota/dataset-summarization-on-joker.md>): Dataset Summarization; Joker; MMDw: 0.164.

- [Dataset Summarization on GMM-2](<https://www.sota2.com/research/sota/dataset-summarization-on-gmm-2.md>): Dataset Summarization; GMM-2; MMDw: 0.0812.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2502.10600>)

[PDF](<https://arxiv.org/pdf/2502.10600.pdf>)
