Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Adaptive Consensus ADMM for Distributed Optimization

About

The alternating direction method of multipliers (ADMM) is commonly used for distributed model fitting problems, but its performance and reliability depend strongly on user-defined penalty parameters. We study distributed ADMM methods that boost performance by using different fine-tuned algorithm parameters on each worker node. We present a O(1/k) convergence rate for adaptive ADMM methods with node-specific parameters, and propose adaptive consensus ADMM (ACADMM), which automatically tunes parameters without user oversight.

Zheng Xu, Gavin Taylor, Hao Li, Mario Figueiredo, Xiaoming Yuan, Tom Goldstein• 2017

Related benchmarks

TaskDatasetResultRank
Elastic net regressionSynthetic Elastic net regression
Runtime (s)623
69
Elastic net regressionsynthetic 2
Iterations57
5
Elastic net regressionMNIST
Iterations14
5
Elastic net regressionnews20
Iterations78
5
Elastic net regressionRCV1
Iterations8
5
Elastic net regressionRealsim
Iterations9
5
Semidefinite programmingHam-9-5-6
Iterations30
5
Sparse Logistic Regressionsynthetic 1
Iterations24
5
Sparse Logistic Regressionsynthetic 2
Iterations114
5
Sparse Logistic RegressionMNIST
Iterations149
5
Showing 10 of 22 rows

Other info

Follow for update