EUVIMEDEUVIMED entdecken
Crossref

Proximal Hamiltonian Monte Carlo

Apratim Shukla, Dootika Vats, Eric C. Chi

IEEE Transactions on Signal Processing · 2026 · 74 · 3170-3181

DOI: 10.1109/tsp.2026.3715313 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Signal Processing, 2026. Verfasst von Apratim Shukla und 2 weiteren Autor:innen. Die Arbeit nennt 77 Literaturangaben. Gefördert von: ANRF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Bayesian formulation of modern day signal processing problems has called for improved Markov chain Monte Carlo (MCMC) sampling algorithms for inference. The need for efficient sampling techniques has become indispensable for high dimensional distributions that often characterize many core signal processing problems, e.g., image denoising, sparse signal recovery, etc. A major issue in building effective sampling strategies, however, is the non-differentiability of the underlying posterior density. Such posteriors are popular in models designed to recover sparse signals. As a result, the use of efficient gradient-based MCMC sampling techniques becomes difficult. We circumvent this problem by proposing a Proximal Hamiltonian Monte Carlo (p-HMC) algorithm, which leverages elements from convex optimization like proximal mappings and Moreau-Yosida (MY) envelopes within Hamiltonian dynamics. We advance the state-of-the-art in non-smooth Hamiltonian Monte Carlo, providing a more accurate log-posterior gradient approximation without increasing computational cost. A chief contribution of this work is the theoretical analysis of p-HMC. We provide conditions for geometric ergodicity of the underlying HMC chain. Further, we propose guidance on choosing the regularization parameter in the MY-envelope. We demonstrate p-HMC’s efficiency over other MCMC algorithms on benchmark problems of logistic regression, low-rank matrix estimation, and compressed sensing.

Förderung

  • ANRF

Zitieren

APA

Shukla, A., Vats, D., & Chi, E. C. (2026). Proximal Hamiltonian Monte Carlo. IEEE Transactions on Signal Processing, 74, 3170-3181. https://doi.org/10.1109/tsp.2026.3715313

Vancouver

Shukla A, Vats D, Chi EC. Proximal Hamiltonian Monte Carlo. IEEE Transactions on Signal Processing. 2026;74:3170-3181. doi:10.1109/tsp.2026.3715313

BibTeX
@article{Shukla2026,
  author = {Shukla, Apratim and Vats, Dootika and Chi, Eric C.},
  title = {Proximal Hamiltonian Monte Carlo},
  journal = {IEEE Transactions on Signal Processing},
  year = {2026},
  volume = {74},
  pages = {3170--3181},
  doi = {10.1109/tsp.2026.3715313}
}
Kostenlos starten →