Imperial College London
Portrait Picture

DrStefanVlaski

Faculty of Engineering, Department of Electrical and Electronic Engineering

Lecturer
 
 
 
//

Contact

 

s.vlaski

 
 
//

Location

 

810Electrical EngineeringSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Hu:2023:10.1109/TIT.2023.3281647,
author = {Hu, P and Bordignon, V and Vlaski, S and Sayed, AH},
doi = {10.1109/TIT.2023.3281647},
journal = {IEEE Transactions on Information Theory},
pages = {6048--6070},
title = {Optimal Aggregation Strategies for Social Learning Over Graphs},
url = {http://dx.doi.org/10.1109/TIT.2023.3281647},
volume = {69},
year = {2023}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Adaptive social learning is a useful tool for studying distributed decision-making problems over graphs. This paper investigates the effect of combination policies on the performance of adaptive social learning strategies. Using large-deviation analysis, it first derives a bound on the steady-state error probability and characterizes the optimal selection for the Perron eigenvectors of the combination policies. It subsequently studies the effect of the combination policy on the transient behavior of the learning strategy by estimating the adaptation time in the low signal-to-noise ratio regime. In the process, it is discovered that, interestingly, the influence of the combination policy on the transient behavior is insignificant, and thus it is more critical to employ policies that enhance the steady-state performance. The theoretical conclusions are illustrated by means of computer simulations.
AU  - Hu,P
AU  - Bordignon,V
AU  - Vlaski,S
AU  - Sayed,AH
DO  - 10.1109/TIT.2023.3281647
EP  - 6070
PY  - 2023///
SN  - 0018-9448
SP  - 6048
TI  - Optimal Aggregation Strategies for Social Learning Over Graphs
T2  - IEEE Transactions on Information Theory
UR  - http://dx.doi.org/10.1109/TIT.2023.3281647
VL  - 69
ER  -
Download