mlsanityData to 14 Oct 2026

Momentum SVGD-EM for Accelerated Maximum Marginal Likelihood Estimation

arXiv 2603.08676 ↗First posted 9 Mar 2026stat.ML cs.LG stat.COcited by 0

Not on the map. Why?

Abstract (arXiv)

Maximum marginal likelihood estimation (MMLE) can be formulated as the optimization of a free energy functional. From this viewpoint, the Expectation-Maximisation (EM) algorithm admits a natural interpretation as a coordinate descent method over the joint space of model parameters and probability measures. Recently, a significant body of work has adopted this perspective, leading to interacting particle algorithms for MMLE. In this paper, we propose an accelerated version of one such procedure, based on Stein variational gradient descent (SVGD), by introducing Nesterov acceleration in both the parameter updates and in the space of probability measures. The resulting method, termed Momentum SVGD-EM, consistently accelerates convergence in terms of required iterations across various tasks of increasing difficulty, demonstrating effectiveness in both low- and high-dimensional settings.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 14 of its 35 references that are in the map. First posted on arXiv 9 Mar 2026.

Builds on

14 of 35 references.

References · 14

  1. 2013Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distancescited by 2,316
  2. 2016Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithmcited by 507
  3. 2022Statistical exploration of the Manifold Hypothesiscited by 29
  4. 2022Particle algorithms for maximum likelihood training of latent variable modelscited by 26
  5. 2023Interacting Particle Langevin Algorithm for Maximum Marginal Likelihood Estimationcited by 17
  6. 2021Algorithmic Instabilities of Accelerated Gradient Descentcited by 16
  7. 2024Error bounds for particle gradient descent, and extensions of the log-Sobolev and Talagrand inequalitiescited by 13
  8. 2023Momentum Particle Maximum Likelihoodcited by 12
  9. 2023Tuning-Free Maximum Likelihood Training of Latent Variable Models via Coin Bettingcited by 11
  10. 2024Proximal Interacting Particle Langevin Algorithmscited by 8
  11. 2024Kinetic Interacting Particle Langevin Monte Carlocited by 7
  12. 2025Accelerated Stein Variational Gradient Flowcited by 6
2 more
  1. 2025Learning Latent Energy-Based Models via Interacting Particle Langevin Dynamicscited by 5
  2. 2025Training Latent Diffusion Models with Interacting Particle Algorithmscited by 4

Built on by

Not cited yet.