Machine-Learning to Trust

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Spiegler, Ran
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914158126039040
author Spiegler, Ran
author_facet Spiegler, Ran
contents Can players sustain long-run trust when their equilibrium beliefs are shaped by machine-learning methods that penalize complexity? I study a game in which an infinite sequence of agents with one-period recall decides whether to place trust in their immediate successor. The cost of trusting is state-dependent. Each player's best response is based on a belief about others' behavior, which is a coarse fit of the true population strategy with respect to a partition of relevant contingencies. In equilibrium, this partition minimizes the sum of the mean squared prediction error and a complexity penalty proportional to its size. Relative to symmetric mixed-strategy Nash equilibrium, this solution concept significantly narrows the scope for trust.
format Preprint
id arxiv_https___arxiv_org_abs_2507_10363
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Machine-Learning to Trust
Spiegler, Ran
Theoretical Economics
Can players sustain long-run trust when their equilibrium beliefs are shaped by machine-learning methods that penalize complexity? I study a game in which an infinite sequence of agents with one-period recall decides whether to place trust in their immediate successor. The cost of trusting is state-dependent. Each player's best response is based on a belief about others' behavior, which is a coarse fit of the true population strategy with respect to a partition of relevant contingencies. In equilibrium, this partition minimizes the sum of the mean squared prediction error and a complexity penalty proportional to its size. Relative to symmetric mixed-strategy Nash equilibrium, this solution concept significantly narrows the scope for trust.
title Machine-Learning to Trust
topic Theoretical Economics
url https://arxiv.org/abs/2507.10363