Remote Empirical Coordination

04/18/2020
by   Michail Mylonakis, et al.
0

We apply the framework of imperfect empirical coordination to a two-node setup where the action X of the first node is not observed directly but via L agents who observe independently impaired measurements X̂ of the action. These L agents, using a rate-limited communication that is available to all of them, help the second node to generate the action Y in order to establish the desired coordinated behaviour. When L<∞, we prove that it suffices R_i≥ I(X̂;Ŷ) for at least one agent whereas for L⟶∞, we show that it suffices R_i≥ I(X̂;Ŷ|X) for all agents where Ŷ is a random variable such that X-X̂-Ŷ and p_X,Ŷ(x,y)-p_X,Y(x,y)_TV≤Δ ( Δ is the pre-specified fidelity).

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/18/2020

Remote Source Coding

We apply the framework of imperfect empirical coordination to a two-node...
research
07/15/2019

Empirical Coordination Subject to a Fidelity Criterion

We study the problem of empirical coordination subject to a fidelity cri...
research
10/05/2019

Empirical Coordination with Multiple Descriptions

We extend the framework of empirical coordination to a distributed setup...
research
02/16/2018

Learning multiagent coordination in the absence of communication channels

In this work, we develop a reinforcement learning protocol for a multiag...
research
07/01/2022

Average submodularity of maximizing anticoordination in network games

We consider the control of decentralized learning dynamics for agents in...
research
02/24/2017

Scalable Multiagent Coordination with Distributed Online Open Loop Planning

We propose distributed online open loop planning (DOOLP), a general fram...
research
01/25/2022

SOL: Safe On-Node Learning in Cloud Platforms

Cloud platforms run many software agents on each server node. These agen...

Please sign up or login with your details

Forgot password? Click here to reset