Being Considerate as a Pathway Towards Pluralistic Alignment for Agentic AI

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Alamdari, Parand A., Klassen, Toryn Q., Icarte, Rodrigo Toro, McIlraith, Sheila A.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916485351342080
author Alamdari, Parand A.
Klassen, Toryn Q.
Icarte, Rodrigo Toro
McIlraith, Sheila A.
author_facet Alamdari, Parand A.
Klassen, Toryn Q.
Icarte, Rodrigo Toro
McIlraith, Sheila A.
contents Pluralistic alignment is concerned with ensuring that an AI system's objectives and behaviors are in harmony with the diversity of human values and perspectives. In this paper we study the notion of pluralistic alignment in the context of agentic AI, and in particular in the context of an agent that is trying to learn a policy in a manner that is mindful of the values and perspective of others in the environment. To this end, we show how being considerate of the future wellbeing and agency of other (human) agents can promote a form of pluralistic alignment.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Being Considerate as a Pathway Towards Pluralistic Alignment for Agentic AI
Alamdari, Parand A.
Klassen, Toryn Q.
Icarte, Rodrigo Toro
McIlraith, Sheila A.
Artificial Intelligence
Computers and Society
Machine Learning
Pluralistic alignment is concerned with ensuring that an AI system's objectives and behaviors are in harmony with the diversity of human values and perspectives. In this paper we study the notion of pluralistic alignment in the context of agentic AI, and in particular in the context of an agent that is trying to learn a policy in a manner that is mindful of the values and perspective of others in the environment. To this end, we show how being considerate of the future wellbeing and agency of other (human) agents can promote a form of pluralistic alignment.
title Being Considerate as a Pathway Towards Pluralistic Alignment for Agentic AI
topic Artificial Intelligence
Computers and Society
Machine Learning
url https://arxiv.org/abs/2411.10613