A random scaling statistic based on averaged AI-SSQP iterates is asymptotically pivotal for constrained stochastic optimization, enabling matrix-free online confidence intervals.
(c) The first result follows from the independence among {St,j}j, and the second result is from (Na and Mahoney, 2025, Corollary 5.4)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Online Statistical Inference of Constrained Stochastic Optimization via Random Scaling
A random scaling statistic based on averaged AI-SSQP iterates is asymptotically pivotal for constrained stochastic optimization, enabling matrix-free online confidence intervals.