Multi-objective Reinforcement Learning: A Tool for Pluralistic Alignment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Vamplew, Peter, Hayes, Conor F, Foale, Cameron, Dazeley, Richard, Harland, Hadassah
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916439765549056
author Vamplew, Peter
Hayes, Conor F
Foale, Cameron
Dazeley, Richard
Harland, Hadassah
author_facet Vamplew, Peter
Hayes, Conor F
Foale, Cameron
Dazeley, Richard
Harland, Hadassah
contents Reinforcement learning (RL) is a valuable tool for the creation of AI systems. However it may be problematic to adequately align RL based on scalar rewards if there are multiple conflicting values or stakeholders to be considered. Over the last decade multi-objective reinforcement learning (MORL) using vector rewards has emerged as an alternative to standard, scalar RL. This paper provides an overview of the role which MORL can play in creating pluralistically-aligned AI.
format Preprint
id arxiv_https___arxiv_org_abs_2410_11221
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-objective Reinforcement Learning: A Tool for Pluralistic Alignment
Vamplew, Peter
Hayes, Conor F
Foale, Cameron
Dazeley, Richard
Harland, Hadassah
Machine Learning
Artificial Intelligence
Reinforcement learning (RL) is a valuable tool for the creation of AI systems. However it may be problematic to adequately align RL based on scalar rewards if there are multiple conflicting values or stakeholders to be considered. Over the last decade multi-objective reinforcement learning (MORL) using vector rewards has emerged as an alternative to standard, scalar RL. This paper provides an overview of the role which MORL can play in creating pluralistically-aligned AI.
title Multi-objective Reinforcement Learning: A Tool for Pluralistic Alignment
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.11221