The Outcomes of Generative AI Are Exactly the Nash Equilibria of a Non-potential Game

Boualem Djehiche, Tembine Hamidou

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

In this article we show that the asymptotic outcomes of both shallow and deep neural networks such as those used in BloombergGPT to generate economic time series are exactly the Nash equilibria of a non-potential game. We then design and analyze deep neural network algorithms that converge to these equilibria. The methodology is extended to federated deep neural networks between clusters of regional servers and on-device clients. Finally, the variational inequalities behind large language models including encoder-decoder related transformers are established.

Original languageEnglish (US)
Title of host publicationStudies in Systems, Decision and Control
PublisherSpringer Science and Business Media Deutschland GmbH
Pages57-80
Number of pages24
DOIs
StatePublished - 2024

Publication series

NameStudies in Systems, Decision and Control
Volume531
ISSN (Print)2198-4182
ISSN (Electronic)2198-4190

Keywords

  • Activation function
  • Deep learning
  • Nash equilibrium
  • Neural network
  • potential game
  • Stackelberg game
  • γ-averaging

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Control and Systems Engineering
  • Automotive Engineering
  • Social Sciences (miscellaneous)
  • Economics, Econometrics and Finance (miscellaneous)
  • Control and Optimization
  • Decision Sciences (miscellaneous)

Fingerprint

Dive into the research topics of 'The Outcomes of Generative AI Are Exactly the Nash Equilibria of a Non-potential Game'. Together they form a unique fingerprint.

Cite this