Saved in:
Bibliographic Details
Main Authors: Sierra, Carles, Osman, Nardine, Noriega, Pablo, Sabater-Mir, Jordi, Perelló, Antoni
Format: Preprint
Published: 2021
Subjects:
Online Access:https://arxiv.org/abs/2110.09240
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914669087686656
author Sierra, Carles
Osman, Nardine
Noriega, Pablo
Sabater-Mir, Jordi
Perelló, Antoni
author_facet Sierra, Carles
Osman, Nardine
Noriega, Pablo
Sabater-Mir, Jordi
Perelló, Antoni
contents principles that should govern autonomous AI systems. It essentially states that a system's goals and behaviour should be aligned with human values. But how to ensure value alignment? In this paper we first provide a formal model to represent values through preferences and ways to compute value aggregations; i.e. preferences with respect to a group of agents and/or preferences with respect to sets of values. Value alignment is then defined, and computed, for a given norm with respect to a given value through the increase/decrease that it results in the preferences of future states of the world. We focus on norms as it is norms that govern behaviour, and as such, the alignment of a given system with a given value will be dictated by the norms the system follows.
format Preprint
id arxiv_https___arxiv_org_abs_2110_09240
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle Value alignment: a formal approach
Sierra, Carles
Osman, Nardine
Noriega, Pablo
Sabater-Mir, Jordi
Perelló, Antoni
Artificial Intelligence
principles that should govern autonomous AI systems. It essentially states that a system's goals and behaviour should be aligned with human values. But how to ensure value alignment? In this paper we first provide a formal model to represent values through preferences and ways to compute value aggregations; i.e. preferences with respect to a group of agents and/or preferences with respect to sets of values. Value alignment is then defined, and computed, for a given norm with respect to a given value through the increase/decrease that it results in the preferences of future states of the world. We focus on norms as it is norms that govern behaviour, and as such, the alignment of a given system with a given value will be dictated by the norms the system follows.
title Value alignment: a formal approach
topic Artificial Intelligence
url https://arxiv.org/abs/2110.09240