New research details Stochastic Mirror Descent for dependent observations

A new research paper explores the Stochastic Mirror Descent (SMD) algorithm, focusing on its convergence properties when dealing with dependent observations rather than the typical independent and identically distributed (i.i.d.) data. The study introduces a novel "forgetting" condition to handle sequences with geometric or polynomial dependencies. This framework yields explicit finite-sample accuracy bounds for averaged SMD iterates, achieving standard rates in different error regimes. The paper demonstrates the applicability of this generalized approach to high-dimensional linear regression with dependent regressors and to the nonparametric estimation of functional autoregressive models. AI

IMPACT This research could improve the robustness of optimization algorithms in machine learning when dealing with real-world, non-i.i.d. data.

RANK_REASON The cluster contains a research paper detailing a new algorithm and its applications. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv stat.ML →

• arXiv
• functional autoregressive model
• linear regression

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →