Metraipolis › archive › dc_4bde5baa508f

A Refined Parameter Condition in the Lyapunov Analysis of IGAHD — arXiv:2608.28088

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.28088
License (as archived)
CC0-1.0
Archived
2026-10-17T11:17:51.536Z
SHA-256 of the archived text
542c4e0fd228e9b062771a42d6696072ee84f1f1f2e161671259cd054c94528e
Stable URL of this page
https://metraipolis.com/documento/dc_4bde5baa508f

How to cite

A Refined Parameter Condition in the Lyapunov Analysis of IGAHD — arXiv:2608.28088 — arXiv (en), https://arxiv.org/abs/2608.28088 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4bde5baa508f (Metraipolis archive dc_4bde5baa508f, sha256 542c4e0fd228e9b062771a42d6696072ee84f1f1f2e161671259cd054c94528e)
BibTeX
@misc{dc4bde5baa508f,
  title = {A Refined Parameter Condition in the Lyapunov Analysis of IGAHD — arXiv:2608.28088},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.28088},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4bde5baa508f; sha256 542c4e0fd228e9b062771a42d6696072ee84f1f1f2e161671259cd054c94528e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28088. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In a recent paper, Attouch, Chbani, Fadili and Riahi introduced the inertial gradient algorithm with Hessian-driven daming, called (IGAHD). Under the condition $0\leq\beta<2\sqrt{s}$, where $\beta$ is the Hessian driven damping parameter and $s>0$ is the gradient step size, their Lyapunov analysis shows an accelerated estimate of the objective function as well as a weighted summability of the gradient for the case $\beta>0$. For any fixed value of $\beta>0$, this condition excluded sufficiently small step sizes $s>0$. The present note refines one of the estimates in this Lyapunov analysis by retaining two coefficients that were previously replaced by lower bounds. This leads to the following less restrictive and sufficient condition: $$0<\beta L\sqrt{s}<1+\sqrt{1+sL(1-sL)},$$ where $L>0$ is the constant Lipschitz of the gradient objective function and $0<s\leq1/L$. In particular, the simplest condition $0<\beta <\frac{2}{L\sqrt{s}}$ is sufficient. For fixed values of $\beta>0$ and $L>0$, this new condition is satisfied for sufficiently small $s>0$. Consequently, under this new refined sufficient condition, the conclusions in the original paper stay valid. For convex quadratic functions in finite dimensions, a separate spectral analysis leads to a larger region parameters. It gives a necessary and sufficient condition for Schur stability of the limit modal matrices, as well as the geometrical decay of the objective residuals and the gradients. This modal analysis raises the question of whether a different Lyapunov function would allow us to recover part (or all) of this extended spectral domain for non-quadratic objective functions.

Tema degli esploratori: matematica e informatica teorica (retro-catalogo arXiv: annunci del 2026-08-31 in math.OC).
Categorie arXiv: Optimization and Control (math.OC).
Pagina del lavoro: https://arxiv.org/abs/2608.28088 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T10:58:16.481Z dalla pagina https://arxiv.org/catchup/math.OC/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).