TY - JOUR
T1 - Stochastic Gradient-Based Distributed Bayesian Estimation in Cooperative Sensor Networks
AU - Cadena, Jose
AU - Ray, Priyadip
AU - Chen, Hao
AU - Soper, Braden
AU - Rajan, Deepak
AU - Yen, Anton
AU - Goldhahn, Ryan
N1 - Cadena, Jose; Ray, Priyadip; Chen, Hao; Soper, Braden; Rajan, Deepak; Yen, Anton; and Goldhahn, Ryan. (2021). "Stochastic Gradient-Based Distributed Bayesian Estimation in Cooperative Sensor Networks". IEEE Transactions on Signal Processing, 69, 1713-1724. https://doi.org/10.1109/TSP.2021.3058765
PY - 2021/2/11
Y1 - 2021/2/11
N2 - Distributed Bayesian inference provides a full quantification of uncertainty offering numerous advantages over point estimates that autonomous sensor networks are able to exploit. However, fully-decentralized Bayesian inference often requires large communication overheads and low network latency, resources that are not typically available in practical applications. In this paper, we propose a decentralized Bayesian inference approach based on stochastic gradient Langevin dynamics, which produces full posterior distributions at each of the nodes with significantly lower communication overhead. We provide analytical results on convergence of the proposed distributed algorithm to the centralized posterior, under typical network constraints. We also provide extensive simulation results to demonstrate the validity of the proposed approach.
AB - Distributed Bayesian inference provides a full quantification of uncertainty offering numerous advantages over point estimates that autonomous sensor networks are able to exploit. However, fully-decentralized Bayesian inference often requires large communication overheads and low network latency, resources that are not typically available in practical applications. In this paper, we propose a decentralized Bayesian inference approach based on stochastic gradient Langevin dynamics, which produces full posterior distributions at each of the nodes with significantly lower communication overhead. We provide analytical results on convergence of the proposed distributed algorithm to the centralized posterior, under typical network constraints. We also provide extensive simulation results to demonstrate the validity of the proposed approach.
KW - Bayes methods
KW - Monte Carlo methods
KW - autonomous systems
KW - distributed algorithms
KW - network theory(graphs)
KW - statistical learning
UR - https://scholarworks.boisestate.edu/electrical_facpubs/495
UR - https://doi.org/10.1109/TSP.2021.3058765
M3 - Article
JO - Electrical and Computer Engineering Faculty Publications and Presentations
JF - Electrical and Computer Engineering Faculty Publications and Presentations
ER -