elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Impressum | Datenschutz | Kontakt | English
Schriftgröße: [-] Text [+]

The Moral Choice Machine

Schramowski, Patrick und Turan, Cigdem und Jentzsch, Sophie Freya und Rothkopf, Constantin und Kersting, Kristian (2020) The Moral Choice Machine. Frontiers in Artificial Intelligence, 3. Frontiers Research Foundation. doi: 10.3389/frai.2020.00036. ISSN 2624-8212.

[img] PDF - Verlagsversion (veröffentlichte Fassung)
958kB

Offizielle URL: https://www.frontiersin.org/articles/10.3389/frai.2020.00036/full

Kurzfassung

Allowing machines to choose whether to kill humans would be devastating for world peace and security. But how do we equip machines with the ability to learn ethical or even moral choices? In this study, we show that applying machine learning to human texts can extract deontological ethical reasoning about “right” and “wrong” conduct. We create a template list of prompts and responses, such as “Should I [action]?”, “Is it okay to [action]?”, etc. with corresponding answers of “Yes/no, I should (not).” and "Yes/no, it is (not)." The model's bias score is the difference between the model's score of the positive response (“Yes, I should”) and that of the negative response (“No, I should not”). For a given choice, the model's overall bias score is the mean of the bias scores of all question/answer templates paired with that choice. Specifically, the resulting model, called the Moral Choice Machine (MCM), calculates the bias score on a sentence level using embeddings of the Universal Sentence Encoder since the moral value of an action to be taken depends on its context. It is objectionable to kill living beings, but it is fine to kill time. It is essential to eat, yet one might not eat dirt. It is important to spread information, yet one should not spread misinformation. Our results indicate that text corpora contain recoverable and accurate imprints of our social, ethical and moral choices, even with context information. Actually, training the Moral Choice Machine on different temporal news and book corpora from the year 1510 to 2008/2009 demonstrate the evolution of moral and ethical choices over different time periods for both atomic actions and actions with context information. By training it on different cultural sources such as the Bible and the constitution of different countries, the dynamics of moral choices in culture, including technology are revealed. That is the fact that moral biases can be extracted, quantified, tracked, and compared across cultures and over time.

elib-URL des Eintrags:https://elib.dlr.de/137728/
Dokumentart:Zeitschriftenbeitrag
Titel:The Moral Choice Machine
Autoren:
AutorenInstitution oder E-Mail-AdresseAutoren-ORCID-iDORCID Put Code
Schramowski, PatrickDarmstadt University of TechnologyNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Turan, CigdemDarmstadt University of TechnologyNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Jentzsch, Sophie FreyaSophie.Jentzsch (at) dlr.dehttps://orcid.org/0000-0001-6217-8814NICHT SPEZIFIZIERT
Rothkopf, ConstantinDarmstadt University of TechnologyNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Kersting, KristianDarmstadt University of TechnologyNICHT SPEZIFIZIERTNICHT SPEZIFIZIERT
Datum:20 Mai 2020
Erschienen in:Frontiers in Artificial Intelligence
Referierte Publikation:Ja
Open Access:Ja
Gold Open Access:Ja
In SCOPUS:Nein
In ISI Web of Science:Ja
Band:3
DOI:10.3389/frai.2020.00036
Verlag:Frontiers Research Foundation
ISSN:2624-8212
Status:veröffentlicht
Stichwörter:Moral, Natural Language Processing, Machine Learning
HGF - Forschungsbereich:Luftfahrt, Raumfahrt und Verkehr
HGF - Programm:Raumfahrt
HGF - Programmthema:Technik für Raumfahrtsysteme
DLR - Schwerpunkt:Raumfahrt
DLR - Forschungsgebiet:R SY - Technik für Raumfahrtsysteme
DLR - Teilgebiet (Projekt, Vorhaben):R - Vorhaben SISTEC (alt)
Standort: Köln-Porz
Institute & Einrichtungen:Institut für Simulations- und Softwaretechnik > Verteilte Systeme und Komponentensoftware
Institut für Softwaretechnologie
Hinterlegt von: Jentzsch, Sophie Freya
Hinterlegt am:23 Nov 2020 14:00
Letzte Änderung:28 Mär 2023 23:57

Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags

Blättern
Suchen
Hilfe & Kontakt
Informationen
electronic library verwendet EPrints 3.3.12
Gestaltung Webseite und Datenbank: Copyright © Deutsches Zentrum für Luft- und Raumfahrt (DLR). Alle Rechte vorbehalten.