Sample Average Approximation for Stochastic Optimization with Dependent Data: Performance Guarantees and Tractability

TL;DR

This paper extends the theoretical guarantees of sample average approximation (SAA) to dependent data scenarios, providing performance bounds and analyzing stochastic algorithms' behavior under dependence, with applications to various optimization algorithms.

Contribution

It proves SAA's consistency and finite-sample guarantees for dependent data and analyzes stochastic algorithms' error bounds under dependence using monotone operator theory.

Findings

SAA remains consistent with dependent data.

Finite-sample performance bounds are established for SAA.

Stochastic algorithms' errors concentrate around zero under dependence.

Abstract

Sample average approximation (SAA), a popular method for tractably solving stochastic optimization problems, enjoys strong asymptotic performance guarantees in settings with independent training samples. However, these guarantees are not known to hold generally with dependent samples, such as in online learning with time series data or distributed computing with Markovian training samples. In this paper, we show that SAA remains tractable when the distribution of unknown parameters is only observable through dependent instances and still enjoys asymptotic consistency and finite sample guarantees. Specifically, we provide a rigorous probability error analysis to derive $1 - \beta$ confidence bounds for the out-of-sample performance of SAA estimators and show that these estimators are asymptotically consistent. We then, using monotone operator theory, study the performance of a class of stochastic first-order algorithms trained on a dependent source of data. We show that approximation error for these algorithms is bounded and concentrates around zero, and establish deviation bounds for iterates when the underlying stochastic process is $\phi$-mixing. The algorithms presented can be used to handle numerically inconvenient loss functions such as the sum of a smooth and non-smooth function or of non-smooth functions with constraints. To illustrate the usefulness of our results, we present several stochastic versions of popular algorithms such as stochastic proximal gradient descent (S-PGD), stochastic relaxed Peaceman--Rachford splitting algorithms (S-rPRS), and numerical experiment.