Recycled ADMM: Improve Privacy and Accuracy with Less Computation in Distributed Algorithms
This addresses privacy-utility trade-offs in distributed convex optimization for applications like federated learning, though it is incremental as it builds on existing ADMM methods.
The paper tackles the privacy leakage and computational cost in distributed ADMM by proposing Recycled ADMM, which uses linear approximations in even iterations to reduce privacy loss by half and cut computation significantly, with convergence and privacy analysis provided.
Alternating direction method of multiplier (ADMM) is a powerful method to solve decentralized convex optimization problems. In distributed settings, each node performs computation with its local data and the local results are exchanged among neighboring nodes in an iterative fashion. During this iterative process the leakage of data privacy arises and can accumulate significantly over many iterations, making it difficult to balance the privacy-utility tradeoff. In this study we propose Recycled ADMM (R-ADMM), where a linear approximation is applied to every even iteration, its solution directly calculated using only results from the previous, odd iteration. It turns out that under such a scheme, half of the updates incur no privacy loss and require much less computation compared to the conventional ADMM. We obtain a sufficient condition for the convergence of R-ADMM and provide the privacy analysis based on objective perturbation.