Gossip Dual Averaging for Decentralized Optimization of Pairwise Functions

Abstract

In decentralized networks (of sensors, connected objects, etc.), there is an important need for efficient algorithms to optimize a global cost function, for instance to learn a global model from the local data collected by each computing unit. In this paper, we address the problem of decentralized minimization of pairwise functions of the data points, where these points are distributed over the nodes of a graph defining the communication topology of the network. This general problem finds applications in ranking, distance metric learning and graph inference, among others. We propose new gossip algorithms based on dual averaging which aims at solving such problems both in synchronous and asynchronous settings. The proposed framework is flexible enough to deal with constrained and regularized variants of the optimization problem. Our theoretical analysis reveals that the proposed algorithms preserve the convergence rate of centralized dual averaging up to an additive bias term. We present numerical simulations on Area Under the ROC Curve (AUC) maximization and metric learning problems which illustrate the practical interest of our approach.

References in corpus (2)

• A Binary Classification Framework for Two-Stage Multiple Kernel Learning
• Extending Gossip Algorithms to Distributed Estimation of U-Statistics

Cited by in corpus (12)

• D${}^2$: Decentralized Training over Decentralized Data
• Hierarchical Federated Learning through LAN-WAN Orchestration
• Communication-Efficient Local Decentralized SGD Methods
• Decentralized and Model-Free Federated Learning: Consensus-Based Distillation in Function Space
• Asynchronous Accelerated Proximal Stochastic Gradient for Strongly Convex Distributed Finite Sums
• Towards More Efficient Stochastic Decentralized Learning: Faster Convergence and Sparse Communication
• Stochastic Composite Least-Squares Regression with convergence rate O(1/n)
• Dual Averaging is Surprisingly Effective for Deep Learning Optimization
• PPT: A Privacy-Preserving Global Model Training Protocol for Federated Learning in P2P Networks
• Matrix Sketching for Secure Collaborative Machine Learning
• Who started this rumor? Quantifying the natural differential privacy guarantees of gossip protocols
• A Riemannian gossip approach to subspace learning on Grassmann manifold