<p>In this paper, we are interested in finding the global minimizer of a nonsmooth nonconvex unconstrained optimization problem. By combining the discrete consensus-based optimization (CBO) algorithm and the gradient descent method, we develop a novel CBO algorithm with an extra gradient descent scheme evaluated by the forward-difference technique on the function values, where only the objective function values are used in the proposed algorithm. First, we prove that the proposed algorithm can exhibit global consensus in an exponential rate in two senses and possess a unique global consensus point. Second, we evaluate the error estimate between the objective function value on the global consensus point and its global minimum. In particular, as the parameter <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10898_2025_1528_Article_IEq1.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="15" /> </InlineMediaObject> <EquationSource Format="TEX">$\beta $</EquationSource> <EquationSource Format="MATHML"><math> <mi>β</mi> </math></EquationSource> </InlineEquation> tends to <InlineEquation ID="IEq2"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10898_2025_1528_Article_IEq2.gif" Format="GIF" Height="10" Rendition="HTML" Resolution="72" Type="Linedraw" Width="21" /> </InlineMediaObject> <EquationSource Format="TEX">$\infty $</EquationSource> <EquationSource Format="MATHML"><math> <mi>∞</mi> </math></EquationSource> </InlineEquation>, the error converges to zero and the convergence rate is <InlineEquation ID="IEq3"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="10898_2025_1528_Article_IEq3.gif" Format="GIF" Height="33" Rendition="HTML" Resolution="72" Type="Linedraw" Width="66" /> </InlineMediaObject> <EquationSource Format="TEX">$\mathcal {O}\left( \frac{\log \beta }{\beta } \right) $</EquationSource> <EquationSource Format="MATHML"><math> <mrow> <mi mathvariant="script">O</mi> <mfenced close=")" open="("> <mfrac> <mrow> <mo>log</mo> <mi>β</mi> </mrow> <mi>β</mi> </mfrac> </mfenced> </mrow> </math></EquationSource> </InlineEquation>. Third, under some suitable assumptions on the objective function, we provide the number of iterations required for the mean square error in expectation to reach the desired accuracy. It is worth underlining that the theoretical analysis in this paper does not use the mean-field limit. Finally, we illustrate the improved efficiency and promising performance of our novel CBO method through some experiments on several nonconvex benchmark problems and the application to train deep neural networks.</p>

A consensus-based optimization method for nonsmooth nonconvex programs with approximated gradient descent scheme

期刊论文
发表日期：

2025年8月8日

查看全文

Jiazhen Wei,
Fan Wu,
Wei Bian

摘要

In this paper, we are interested in finding the global minimizer of a nonsmooth nonconvex unconstrained optimization problem. By combining the discrete consensus-based optimization (CBO) algorithm and the gradient descent method, we develop a novel CBO algorithm with an extra gradient descent scheme evaluated by the forward-difference technique on the function values, where only the objective function values are used in the proposed algorithm. First, we prove that the proposed algorithm can exhibit global consensus in an exponential rate in two senses and possess a unique global consensus point. Second, we evaluate the error estimate between the objective function value on the global consensus point and its global minimum. In particular, as the parameter $\beta $ $β$ tends to $\infty $ $\infty$ , the error converges to zero and the convergence rate is $\mathcal {O}\left( \frac{\log \beta }{\beta } \right) $ $O (\frac{log β}{β})$ . Third, under some suitable assumptions on the objective function, we provide the number of iterations required for the mean square error in expectation to reach the desired accuracy. It is worth underlining that the theoretical analysis in this paper does not use the mean-field limit. Finally, we illustrate the improved efficiency and promising performance of our novel CBO method through some experiments on several nonconvex benchmark problems and the application to train deep neural networks.