Network formation by reinforcement learning: the long and medium run

DOI10.1016/J.MATHSOCSCI.2004.03.007zbMATH Open1091.91060arXivmath/0404106OpenAlexW2112233006MaRDI QIDQ1763191FDOQ1763191

Authors: Robin Pemantle, Brian Skyrms 

Publication date: 22 February 2005

Published in: Mathematical Social Sciences (Search for Journal in Brave)

Abstract: We investigate a simple stochastic model of social network formation by the process of reinforcement learning with discounting of the past. In the limit, for any value of the discounting parameter, small, stable cliques are formed. However, the time it takes to reach the limiting state in which cliques have formed is very sensitive to the discounting parameter. Depending on this value, the limiting result may or may not be a good predictor for realistic observation times.

Full work available at URL: https://arxiv.org/abs/math/0404106

Recommendations

• A dynamic model of social network formation
• Distributed dynamic reinforcement of efficient outcomes in multiagent coordination and network formation
• Strategic learning and the topology of social networks
• A Noncooperative Model of Network Formation
• Stochastic stability in networks with decay

zbMATH Keywords

Stochastic approximationUrn modelFriedmanExponential timeAbsorptionDiscountMeta-stablePotential wellQuasi-stableStag HuntThree-player gameTrap

Mathematics Subject Classification ID

Social networks; opinion dynamics (91D30)

Cites Work

• Title not available (Why is that?)
• On learning to become a successful loser: A comparison of alternative abstractions of learning processes in the loss domain
• Learning in extensive-form games: Experimental data and simple dynamic models in the intermediate term
• Experience-weighted Attraction Learning in Normal Form Games
• Phase transition in reinforced random walk and RWRE on trees
• Reinforced random walk
• Title not available (Why is that?)
• A dynamic model of social network formation
• On the convergence of reinforcement learning
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Recursive algorithms, urn processes and chaining number of chain recurrent sets
• Vertex-reinforced random walk
• Asymptotics of a matrix valued Markov chain arising in sociology.
• Dynamics of Morse-Smale urn processes
• Attracting edge property for a class of reinforced random walks
• Time to absorption in discounted reinforcement models.

Cited In (9)

• Models of coalition or alliance formation
• Interaction dynamics of two reinforcement learners
• VRRW on complete-like graphs: almost sure behavior
• The statistical properties of the threshold model and the feedback leadership condition
• Dynamic network formation with foresighted agents
• Time to absorption in discounted reinforcement models.
• Wisdom of crowds versus groupthink: learning in groups and in isolation
• Distributed dynamic reinforcement of efficient outcomes in multiagent coordination and network formation
• An infinite stochastic model of social network formation

This page was built for publication: Network formation by reinforcement learning: the long and medium run

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1763191)