Distributed Robust Subspace Recovery

DOI10.1137/17M1131659zbMATH Open1484.68212arXiv1705.09382OpenAlexW3105285841MaRDI QIDQ4685332FDOQ4685332

Authors: Vahan Huroyan, Gilad Lerman 

Publication date: 8 October 2018

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Abstract: We propose distributed solutions to the problem of Robust Subspace Recovery (RSR). Our setting assumes a huge dataset in an ad hoc network without a central processor, where each node has access only to one chunk of the dataset. Furthermore, part of the whole dataset lies around a low-dimensional subspace and the other part is composed of outliers that lie away from that subspace. The goal is to recover the underlying subspace for the whole dataset, without transferring the data itself between the nodes. We first apply the Consensus-Based Gradient method to the Geometric Median Subspace algorithm for RSR. For this purpose, we propose an iterative solution for the local dual minimization problem and establish its r-linear convergence. We then explain how to distributedly implement the Reaper and Fast Median Subspace algorithms for RSR. The proposed algorithms display competitive performance on both synthetic and real data.

Full work available at URL: https://arxiv.org/abs/1705.09382

Recommendations

• Distributed estimation based on multi-hop subspace decomposition
• Robust Recovery of Signals From a Structured Union of Subspaces
• Robust Distributed Estimation Using the Embedded Subgraphs Algorithm
• Fast, robust and non-convex subspace recovery
• Robust and scalable distributed recursive least squares
• Distributed matrix completion and robust factorization
• Subspace Methods for Joint Sparse Recovery
• Distributed reconstruction via alternating direction method
• A robust and efficient algorithm for distributed compressed sensing
• Distributed Recovery of Jointly Sparse Signals Under Communication Constraints

zbMATH Keywords

geometric mediandistributed algorithmsprincipal component analysis (PCA)consensus-based algorithmsrobust subspace recovery (RSR)

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Computational aspects of data analysis and big data (68T09) Large-scale problems in mathematical programming (90C06) Distributed algorithms (68W15)

Cites Work

• Fast, robust and non-convex subspace recovery
• Algorithm 432 [C2]: Solution of the matrix equation AX + XB = C [F4]
• Title not available (Why is that?)
• Distributed optimization and statistical learning via the alternating direction method of multipliers
• Consensus-Based Distributed Total Least Squares Estimation in Ad Hoc Wireless Sensor Networks
• Title not available (Why is that?)
• Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions
• A novel M-estimator for robust PCA
• Convex Analysis
• Consensus in Ad Hoc WSNs With Noisy Links—Part I: Distributed Estimation of Deterministic Signals
• Distributed Sparse Linear Regression
• On the Linear Convergence of the ADMM in Decentralized Consensus Optimization
• Distributed Subgradient Methods for Multi-Agent Optimization
• Title not available (Why is that?)
• Title not available (Why is that?)
• Cooperative distributed multi-agent optimization
• Robust computation of linear models by convex relaxation
• On the Convergence of the Lagged Diffusivity Fixed Point Method in Total Variation Image Restoration
• Regularization techniques for learning with matrices
• Consensus-based distributed support vector machines
• Title not available (Why is that?)
• Diffusion Adaptation Strategies for Distributed Optimization and Learning Over Networks
• Multi-Agent Distributed Optimization via Inexact Consensus ADMM

Cited In (3)

• A novel M-estimator for robust PCA
• Fast, robust and non-convex subspace recovery
• A well-tempered landscape for non-convex robust subspace recovery

Uses Software

• UCI-ml
• Algorithm 432
• FMS

This page was built for publication: Distributed Robust Subspace Recovery

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4685332)