We consider the problem of minimizing a differentiable function with locally Lipschitz continuous gradient over the real determinantal variety, and present a first-order algorithm designed to find stationary points of that problem. This algorithm applies steps of a retraction-free descent method proposed by Schneider and Uschmajew (2015), while taking the numerical rank into account to attempt rank reductions. We prove that this algorithm produces a sequence of iterates the accumulation points of which are stationary, and therefore does not follow the so-called apocalypses described by Levin, Kileel, and Boumal (2022). Moreover, the rank reduction mechanism of this algorithm requires at most one rank reduction attempt per iteration, in contrast with the one of the $\mathrm{P}^2\mathrm{GDR}$ algorithm introduced by Olikier, Gallivan, and Absil (2022) which can require a number of rank reduction attempts equal to the rank of the iterate in the worst-case scenario.
翻译:我们考虑的是将局部Lipschitz连续梯度与真实的决定因素多样性相比的可变函数最小化的问题,并且提出一种旨在找到这一问题固定点的一阶算法。 这种算法采用了Schneider和Uschmajew(2015年)建议的不撤回的下降方法的步骤,同时将数值等级考虑在内以尝试降级。我们证明,这种算法产生了一系列循环,其积累点是固定的,因此不遵循Levin、Kileel和Boumal(2022年)所描述的所谓“末日化 ” 。 此外,这种算法的降级机制要求每次迭代最多要尝试一次降级尝试一次,这与Olikier、Gallivan和Absil(2022年)提出的“$mathrm{P ⁇ 2\mathrm}GDRDR}$算法不同,后者可能需要一系列降级尝试,其降级与最坏情况下的 Iterate等级相等。