Aditya Dave, Andreas A. Malikopoulos
Published 2019-07-28Version 1
In this paper, we analyze a network of agents in a partially nested information structure with a common ancestor. We present the prescription approach applied to different permutations of agents and a structural result for optimal prescriptions of control strategies. We demonstrate the proposed approach through an example that aims at establishing time-invariant domains of the prescriptions without assuming a Linear Quadratic Gaussian problem.
Reinforcement Learning for Decentralized Stochastic Control and Coordination Games
On a General Dynamic Programming Approach for Decentralized Stochastic Control
Decentralized Stochastic Control with a Word-of-Mouth Communication
![QR code for arXiv:1907.12125 [math.OC]](http://oss.arxitics.com/images/favicon.svg)