Id: arxiv/2611.11459 Url: https://arxiv.org/abs/2611.11459 Tags: math.OC Score: 0 Title: From Imperfect Information to Controlled Steps: Stochastic Optimization under Heavy-Tailed Noise and Compression Domain: math.OC Source: arxiv Authors: Abdurakhmon Sadiev Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-11-08 Page: https://delphi.science/r/arxiv/2611.11459 Stochastic optimization relies on information whose statistical quality, availability, and computational cost can vary substantially. Heavy-tailed observations challenge concentration, compression distorts communicated vectors, and generalized smoothness changes how far an algorithm can move. This dissertation studies these difficulties through two linked design decisions: how to construct an informative search vector and how to convert it into a controlled step. We first develop high-probability methods under bounded central $p$-th moments, $p\in(1,2]$, allowing infinite variance when…