Payoff measurement noise in risk-sensitive mean-field-type games

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Payoff measurement noise constitutes a major problem in practical interactive decision-making scenarios. When agent's payoff is erroneously perceived or observed with a noise functional, the response is highly impacted. In this paper, a risk-sensitive approach is proposed for capturing players' behaviors in presence of measurement randomness in mean-field-type games. Equilibrium and optimal strategy systems are established using stochastic maximum principle and dynamic programming principle in infinite dimensions.

Original languageEnglish (US)
Title of host publicationProceedings of the 29th Chinese Control and Decision Conference, CCDC 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3764-3769
Number of pages6
ISBN (Electronic)9781509046560
DOIs
StatePublished - Jul 12 2017
Event29th Chinese Control and Decision Conference, CCDC 2017 - Chongqing, China
Duration: May 28 2017May 30 2017

Publication series

NameProceedings of the 29th Chinese Control and Decision Conference, CCDC 2017

Other

Other29th Chinese Control and Decision Conference, CCDC 2017
CountryChina
CityChongqing
Period5/28/175/30/17

Keywords

  • Dynamic programming
  • Game theory
  • Mean-field
  • Stochastic control
  • Stochastic maximum principle

ASJC Scopus subject areas

  • Decision Sciences (miscellaneous)
  • Control and Optimization

Fingerprint Dive into the research topics of 'Payoff measurement noise in risk-sensitive mean-field-type games'. Together they form a unique fingerprint.

Cite this