Id: arxiv/2607.09097 Url: https://arxiv.org/abs/2607.09097 Tags: math.OC, cs.DS, cs.LG, stat.ML Score: 0 Title: Solving Stochastic Fixed-Point Equations with High Probability Domain: math.OC Source: arxiv Authors: Jelena Diakonikolas Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-07-10 Page: https://delphi.science/r/arxiv/2607.09097 We study stochastic fixed-point equations $\mathbf{T}(\mathbf{x}) = \mathbf{x}$ over normed spaces $(\mathcal{E}, \|\cdot\|)$, where the operator $\mathbf{T}$ is nonexpansive or contractive and is accessed only through unbiased stochastic evaluations with bounded second central moment. Given $ε> 0, δ\in (0, 1)$, the goal is to output $\mathbf{x} \in \mathcal{E}$ such that $\|\mathbf{T}(\mathbf{x}) - \mathbf{x}\| \leq ε$ with probability at least $1-δ$. We introduce VR-GHAL, a variance-reduced gradual Halpern method for quadratically smoothable Banach spaces. The key algorithmic ingredient is…