Communication-efficient Distributed Sparse Linear Discriminant Analysis

Abstract

We propose a communication-efficient distributed estimation method for sparse linear discriminant analysis (LDA) in the high dimensional regime. Our method distributes the data of size $N$ into $m$ machines, and estimates a local sparse LDA estimator on each machine using the data subset of size $N/m$. After the distributed estimation, our method aggregates the debiased local estimators from $m$ machines, and sparsifies the aggregated estimator. We show that the aggregated estimator attains the same statistical rate as the centralized estimation method, as long as the number of machines $m$ is chosen appropriately. Moreover, we prove that our method can attain the model selection consistency under a milder condition than the centralized method. Experiments on both synthetic and real datasets corroborate our theory.

29 pages, 2 figures, 2 tables

References in corpus (6)

• Simultaneous analysis of Lasso and Dantzig selector
• Confidence Intervals and Hypothesis Testing for High-Dimensional Regression
• Optimal Distributed Online Prediction using Mini-Batches
• Improved Distributed Principal Component Analysis
• Distributed Estimation and Inference with Statistical Guarantees
• Communication-efficient sparse regression: a one-shot approach

Cited by in corpus (4)

• Heterogeneity-aware and communication-efficient distributed statistical inference
• Communication-efficient Distributed Estimation and Inference for Transelliptical Graphical Models
• SecureGBM: Secure Multi-Party Gradient Boosting
• A selective review on calibration information from similar studies based on parametric likelihood or empirical likelihood