Diegetic Representation of Feedback in Open Games
From MaRDI portal
Publication:6150152
DOI10.4204/EPTCS.380.9arXiv2206.12338MaRDI QIDQ6150152FDOQ6150152
Authors:
Publication date: 5 March 2024
Published in: Electronic Proceedings in Theoretical Computer Science (Search for Journal in Brave)
Abstract: We improve the framework of open games with agency by showing how the players' counterfactual analysis giving rise to Nash equilibria can be described in the dynamics of the game itself (hence diegetically), getting rid of devices such as equilibrium predicates. This new approach overlaps almost completely with the way gradient-based learners are specified and trained. Indeed, we show feedback propagation in games can be seen as reverse-mode differentiation, with a crucial difference explaining the distinctive character of the phenomenology of non-cooperative games. We outline a functorial construction of arena of games, show players form a subsystem over it, and prove that their `fixpoint behaviours' are Nash equilibria.
Full work available at URL: https://arxiv.org/abs/2206.12338
Recommendations
Cites Work
- Title not available (Why is that?)
- Games with Incomplete Information Played by “Bayesian” Players, I–III Part I. The Basic Model
- Essentials of game theory. A concise, multidisciplinary introduction.
- Compositional game theory
- A compositional treatment of iterated open games
- Higher-order decision theory
- 2-Dimensional Categories
- Morphisms of open games
- Title not available (Why is that?)
- Homotopy type theory: the logic of space
- Double categories of open dynamical systems (extended abstract)
- Categorical foundations of gradient-based learning
Cited In (1)
This page was built for publication: Diegetic Representation of Feedback in Open Games
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6150152)
