Adaptive Consensus ADMM for Distributed Optimization
This work improves distributed model fitting for applications like machine learning by making ADMM more reliable and efficient without manual tuning, though it is incremental as it builds on existing ADMM methods.
The paper tackles the problem of distributed optimization with ADMM by addressing its sensitivity to user-defined penalty parameters, resulting in an adaptive method (ACADMM) that automatically tunes parameters and achieves a O(1/k) convergence rate.
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.