Distributed event-triggered unadjusted Langevin algorithm for Bayesian learning

DOI10.1016/J.AUTOMATICA.2023.111221MaRDI QIDQ6136164zbMATH OpenFDO

Authors Kinjal Bhar, He Bai, Jemin George, Carl Busart

Publication date 28 August 2023

Published in Automatica (Search for Journal in Brave)

zbMATH Keywords

Bayesian learningLangevin dynamicsevent-triggered communicationdistributed multi-agent networks

Mathematics Subject Classification ID

Discrete event control/observation systems (93C65) Multi-agent systems (93A16) Networked control (93B70)

Recommendations

• Distributed Bayesian learning with stochastic natural gradient expectation propagation and the posterior server
• Distributed Bayesian machine learning procedures
• Decentralized Bayesian learning with Metropolis-adjusted Hamiltonian Monte Carlo
• Distributed Variational Bayesian Algorithms Over Sensor Networks
• Hybrid deterministic-stochastic gradient Langevin dynamics for Bayesian learning
• scientific article; zbMATH DE number 7365721
• Fast Convergence Rates for Distributed Non-Bayesian Learning
• Stochastic Event-triggered Variational Bayesian Filtering

Cites work

• scientific article; zbMATH DE number 7626754 (Why is no real title available?)
• An introduction to MCMC for machine learning
• Convergence of Langevin MCMC in KL-divergence
• Distributed Event-Triggered Control for Multi-Agent Systems
• Distributed convex optimization via continuous-time coordination algorithms with discrete-time communication
• Distributed linear parameter estimation: asymptotically efficient adaptive strategies
• Dynamic Triggering Mechanisms for Event-Triggered Control
• Exponential convergence of Langevin distributions and their discrete approximations
• Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality
• Generalized inverse of the Laplacian matrix and some applications
• High-dimensional Bayesian inference via the unadjusted Langevin algorithm
• Improved bounds for discretization of Langevin diffusions: near-optimal rates without convexity
• Logarithmic Sobolev Inequalities
• Nonasymptotic bounds for sampling algorithms without log-concavity
• Nonasymptotic convergence analysis for the unadjusted Langevin algorithm
• Nonasymptotic estimates for stochastic gradient Langevin dynamics under local conditions in nonconvex optimization
• On Stochastic Gradient Langevin Dynamics with Dependent Data Streams: The Fully Nonconvex Case
• SPARQ-SGD: Event-Triggered and Compressed Communication in Decentralized Optimization
• Sampling can be faster than optimization
• Simulating Hamiltonian Dynamics
• The Variational Formulation of the Fokker--Planck Equation
• Theoretical Guarantees for Approximate Sampling from Smooth and Log-Concave Densities
• Variance-Reduced Decentralized Stochastic Optimization With Accelerated Convergence

Cited in

(4)

• Decentralized Bayesian learning with Metropolis-adjusted Hamiltonian Monte Carlo
• Decentralized fused-learner architectures for Bayesian reinforcement learning
• Event-triggered learning
• Stochastic Event-triggered Variational Bayesian Filtering

This page was built for publication: Distributed event-triggered unadjusted Langevin algorithm for Bayesian learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136164)