Diffusion Approximations for Online Principal Component Estimation and Global Convergence

Authors: Chris Junchi Li, Mengdi Wang, Han Liu, Tong Zhang

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical In this paper, we propose to adopt the diffusion approximation tools to study the dynamics of Oja s iteration... We show that the Oja s iteration for the top eigenvector generates a continuous-state discrete-time Markov chain over the unit sphere. We characterize the Oja s iteration in three phases using diffusion approximation and weak convergence tools. Our three-phase analysis further provides a finite-sample error bound for the running estimate, which matches the minimax information lower bound for principal component analysis under the additional assumption of bounded samples.
Researcher Affiliation Collaboration Chris Junchi Li Mengdi Wang Han Liu Princeton University Department of Operations Research and Financial Engineering, Princeton, NJ 08544 {junchil,mengdiw,hanliu}@princeton.edu Tong Zhang Tencent AI Lab Shennan Ave, Nanshan District, Shenzhen, Guangdong Province 518057, China tongzhang@tongzhang-ml.org
Pseudocode No The paper describes the Oja's iteration using mathematical notation (Equation 1.3) but does not provide any explicitly labeled 'Algorithm' or 'Pseudocode' block.
Open Source Code No The paper does not contain any statement about releasing source code or provide a link to a code repository for the methodology described.
Open Datasets No The paper is a theoretical analysis of an algorithm and does not describe experiments performed on specific public datasets for training.
Dataset Splits No The paper is theoretical and does not describe an experimental setup with dataset splits for training, validation, or testing.
Hardware Specification No The paper does not provide any specific details about the hardware used for any computational work, which is consistent with its theoretical focus.
Software Dependencies No The paper does not list any specific software dependencies with version numbers, such as programming languages, libraries, or frameworks used for implementation or analysis.
Experiment Setup No The paper does not detail specific experimental setup parameters such as hyperparameter values, training configurations, or system-level settings, as it is a theoretical work.