arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.05893math.OC
收藏

无目标函数的多目标优化:Adagrad类算法的收敛速率与性能

Objective-Function Free Multi-Objective Optimization: Rate of Convergence and Performance of an Adagrad-like algorithm

Università degli Studi di Firenze(佛罗伦萨大学);Technische Universität Ilmenau(埃尔福特工业大学);ISTI–CNR, Italy(意大利国家研究委员会信息系统研究所)

Marianna De Santis, Gabriele Eichfelder, Margherita Porcelli

AI总结:

提出一种不依赖目标函数值、仅使用公共下降方向的Adagrad类多目标优化算法,具有自适应步长,收敛速率为O(1/√(k+1)),在无约束多目标问题和多任务学习中验证了有效性。

摘要:

我们提出了一种用于多目标无约束优化的Adagrad类算法,该算法仅依赖于计算公共下降方向。与经典的多目标局部优化算法不同,我们的方法不依赖支配性质来接受新迭代点,从而实现了灵活且无函数依赖的优化框架。新点通过自适应步长获得,该步长既不需要知道Lipschitz常数,也不需要线搜索过程。分析了收敛速率,并证明相对于公共下降方向的范数,收敛速率为$\mathcal{O}(1 / \sqrt{ k+1})$。该方法在一大类无约束多目标问题和简单的多任务学习实例上进行了广泛验证,并与一阶线搜索算法进行了比较。此外,我们初步研究了在噪声多目标设置下的行为,突出了该方法的鲁棒性。

英文摘要:

We propose an Adagrad-like algorithm for multi-objective unconstrained optimization that relies on the computation of a common descent direction only. Unlike classical local algorithms for multi-objective optimization, our approach does not rely on the dominance property to accept new iterates, which allows for a flexible and function-free optimization framework. New points are obtained using an adaptive stepsize that does not require neither knowledge of Lipschitz constants nor the use of line search procedures. The rate of convergence is analyzed and is shown to be $\mathcal{O}(1 / \sqrt{ k+1})$ with respect to the norm of the common descent direction. The method is extensively validated on a broad class of unconstrained multi-objective problems and simple multi-task learning instances, and compared against a first-order line search algorithm. Additionally, we present a preliminary study of the behavior under noisy multi-objective settings, highlighting the robustness of the method.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑