Online supervised learning with distributed features over multiagent system
Summary: Most current online distributed machine learning algorithms have been studied in a data-parallel architecture among agents in networks. We study online distributed machine learning from a different perspective, where the features about the same samples are observed by multiple agents that wish to collaborate but do not exchange the raw data with each other. We propose a distributed feature online gradient descent algorithm and prove that local solution converges to the global minimizer with a sublinear rate \(\mathrm{O}(\sqrt{2T})\). Our algorithm does not require exchange of the primal data or even the model parameters between agents. Firstly, we design an auxiliary variable, which implies the information of the global features, and estimate at each agent by dynamic consensus method. Then, local parameters are updated by online gradient descent method based on local data stream. Simulations illustrate the performance of the proposed algorithm.
- Online learning over a decentralized network through ADMM
- A distributed conjugate gradient online learning method over networks
- Distributed multi-task classification: a decentralized online learning approach
- Consensus-based distributed support vector machines
- Kernel-based online gradient descent using distributed approach
- A novel clustering method based on quasi-consensus motions of dynamical multiagent systems
- Adaptation, learning, and optimization over networks
- Consensus of linear multi-agent systems with persistent disturbances via distributed output feedback
- Dictionary Learning Over Distributed Models
- Discrete-time dynamic average consensus
- Distributed Constrained Optimization by Consensus-Based Primal-Dual Perturbation Method
- Distributed optimization and statistical learning via the alternating direction method of multipliers
- Distributed Subgradient Methods for Multi-Agent Optimization
- Dual decomposition for multi-agent distributed optimization with coupling constraints
- EXTRA: an exact first-order algorithm for decentralized consensus optimization
- scientific article; zbMATH DE number 611919 (Why is no real title available?)
- scientific article; zbMATH DE number 7385968 (Why is no real title available?)
- Jordan form-based algebraic conditions for controllability of multiagent systems under directed graphs
- Multi-Agent Distributed Optimization via Inexact Consensus ADMM
- On almost controllability of dynamical complex networks with noises
- On the Linear Convergence of the ADMM in Decentralized Consensus Optimization
- Online learning and online convex optimization
- Robust time-varying formation design for multiagent systems with disturbances: extended-state-observer method
- SGDLibrary: a MATLAB library for stochastic optimization algorithms
- Supervised Learning Under Distributed Features
This page was built for publication: Online supervised learning with distributed features over multiagent system
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2225196)
