Regularized Sample Average Approximation for High-Dimensional Stochastic Optimization Under Low-Rankness

DOI10.1007/S10898-022-01206-3arXiv1904.03453MaRDI QIDQ6316795FDOQ6316795

Authors: Hung-yi Lee, Charles Hernandez, Hongcheng Liu 

Publication date: 6 April 2019

Abstract: This paper concerns a high-dimensional stochastic programming problem of minimizing a function of expected cost with a matrix argument. To this problem, one of the most widely applied solution paradigms is the sample average approximation (SAA), which uses the average cost over sampled scenarios as a surrogate to approximate the expected cost. Traditional SAA theories require the sample size to grow rapidly when the problem dimensionality increases. Indeed, for a problem of optimizing over a $p$-by-$p$ matrix, the sample complexity of the SAA is given by $ildeO(1)cdotfrac{p}^{2}epsilo{n}^{2}cdotpolylog(frac1epsilon)$ to achieve an $epsilon$-suboptimality gap, for some poly-logarithmic function $polylog(,cdot,)$ and some quantity $ildeO(1)$ independent of dimensionality $p$ and sample size $n$. In contrast, this paper considers a regularized SAA (RSAA) with a low-rankness-inducing penalty. We demonstrate that the sample complexity of RSAA is $ildeO(1)cdotfracpepsilo{n}^{3}cdotpolylog(p,,frac1epsilon)$, which is almost linear in $p$ and thus indicates a substantially lower dependence on dimensionality. Therefore, RSAA can be more advantageous than SAA especially for larger scale and higher dimensional problems. Due to the close correspondence between stochastic programming and statistical learning, our results also indicate that high-dimensional low-rank matrix recovery is possible generally beyond a linear model, even if the common assumption of restricted strong convexity is completely absent.

Mathematics Subject Classification ID

Approximation methods and heuristics in mathematical programming (90C59) Stochastic programming (90C15)

This page was built for publication: Regularized Sample Average Approximation for High-Dimensional Stochastic Optimization Under Low-Rankness

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6316795)