PreprintCC BY 4.0arXiv
Stochastic Optimization and Data Science
Abstract
This paper aims to motivate stochastic optimization problems from a statistical perspective and a statistical learning perspective, where the goal is to maximize the log-likelihood or minimize the population risk. We briefly describe the two main approaches: offline (Monte Carlo / Sample Average Approximation) and online (Stochastic Approximation) approaches -- to solve the expectation minimization problems.
Related
- Spatio-Temporal Stochastic Interventions for Causal Inference in Climate Science
- Stochastic Optimization of Tree Tensor Networks
- Random tilts to find stationary points in stochastic convex optimization
- Gradient-Free Methods for Stochastic Convex Optimization with Stochastic Functional Constraints
- Accelerated Stochastic Zeroth-Order Quasar-Convex Optimization
BibTeX
@misc{avetisyan2026stochastic,
title = {{Stochastic Optimization and Data Science}},
author = {Arutyun Avetisyan and Darina Dvinskikh and Alexander Gasnikov and Vladimir Temlyakov and Nazarii Tupitsa and Denis Turdakov},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2605.16875},
eprint = {2605.16875},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.16875},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Stochastic Optimization and Data Science
AU - Arutyun Avetisyan
AU - Darina Dvinskikh
AU - Alexander Gasnikov
AU - Vladimir Temlyakov
AU - Nazarii Tupitsa
AU - Denis Turdakov
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2605.16875
UR - https://arxiv.org/abs/2605.16875
AB - This paper aims to motivate stochastic optimization problems from a statistical perspective and a statistical learning perspective, where the goal is to maximize the log-likelihood or minimize the population risk. We briefly describe the two main approaches: offline (Monte Carlo / Sample Average Approximation) and online (Stochastic Approximation) approaches -- to solve the expectation minimization problems.
ER -