Sequential decomposition of repeated games with asymmetric information and dependent states | IEEE Conference Publication | IEEE Xplore

Sequential decomposition of repeated games with asymmetric information and dependent states


Abstract:

We consider a finite horizon repeated game with N selfish players who observe their types privately and take actions, which are publicly observed. Their actions and types...Show More

Abstract:

We consider a finite horizon repeated game with N selfish players who observe their types privately and take actions, which are publicly observed. Their actions and types jointly determine their instantaneous rewards. In each period, players jointly observe actions of each other with delay 1, and private observations of the state of the system, and get an instantaneous reward which is a function of the state and everyone's actions. The players' types are static and are potentially correlated among players. An appropriate notion of equilibrium for such games is Perfect Bayesian Equilibrium (PBE) which consists of a strategy and a belief profile of the players which is coupled across time and as a result, the complexity of finding such equilibria grows double-exponentially in time. We present a sequential decomposition methodology to compute structured perfect Bayesian equilibria (SPBE) of this game, introduced in [1], where equilibrium policy of a player is a function of a common belief and a private state. This methodology computes SPBE in linear time. In general, the SPBE of the game problem exhibit signaling behavior, i.e. players' actions reveal part of their private information that is payoff relevant to other players.
Date of Conference: 10-12 July 2019
Date Added to IEEE Xplore: 29 August 2019
ISBN Information:

ISSN Information:

Conference Location: Philadelphia, PA, USA

I. Introduction

Information asymmetry among strategic agents is an important topic, which has seen some very influential works such as [2] and [3]. Akerlof in [2] and Spence in [3] modeled a market of cars and a job market, respectively, as instances of information asymmetry in a game, and show interesting behavior of strategic agents derived from these models. Specifically, Akerlof in [2] showed that in a market of cars, where the quality of car is known only to the seller, lower prices can drive out good cars from the market. Spence in [3] showed that in equilibrium in a job market, a candidate can ‘signal’ her higher productivity to a potential employer by opting for higher education credentials. While these works showed very interesting and relevant phenomena for static information asymmetry, in the real world however, there exists many such, and even more complicated decision making scenarios which involves strategic decision makers with dynamically evolving information asymmetry. Some instances of such systems include: (a) in cyber-physical systems, many cyber and physical devices are connected to each other which have different information and they make a decision to optimize their performance objectives; (b) in a wind energy market a wind energy producer observes its own wind production privately and publicly observes the output of other producers which also determine the prices, and its objective is to generate output that maximizes its revenue; (c) in a social network, people have private opinions about a topic and also publicly observe actions of others, based on which they make a decision to maximize their utility. All such scenarios can be modeled as a dynamic game of asymmetric information

Sometimes also referred to as dynamic games of incomplete/imperfect information.

[4]. Such problems are gaining more interest with applications such as alpha-go by Deepmind [5] for the symmetric information game Go, and for asymmetric/imperfect information games such as Texas Hold'em in [6].

Contact IEEE to Subscribe

References

References is not available for this document.