Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Lerner, Osher
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2404.17725
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917652211957760
author Lerner, Osher
author_facet Lerner, Osher
contents This paper expands on existing learned models of human behavior via a measured step in structured irrationality. Specifically, by replacing the suboptimality constant $β$ in a Boltzmann rationality model with a function over states $β(s)$, we gain natural expressivity in a computationally tractable manner. This paper discusses relevant mathematical theory, sets up several experimental designs, presents limited preliminary results, and proposes future investigations.
format Preprint
id arxiv_https___arxiv_org_abs_2404_17725
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Boltzmann State-Dependent Rationality
Lerner, Osher
Artificial Intelligence
Robotics
This paper expands on existing learned models of human behavior via a measured step in structured irrationality. Specifically, by replacing the suboptimality constant $β$ in a Boltzmann rationality model with a function over states $β(s)$, we gain natural expressivity in a computationally tractable manner. This paper discusses relevant mathematical theory, sets up several experimental designs, presents limited preliminary results, and proposes future investigations.
title Boltzmann State-Dependent Rationality
topic Artificial Intelligence
Robotics
url https://arxiv.org/abs/2404.17725