Engineering and Reverse-Engineering Morality
Workshop at Cogsci 2021
July 26 (Virtual)
Recent years have witnessed a burst of progress on building formal models of moral decision-making. In psychology, neuroscience and philosophy, the goal has been to “reverse-engineer” the principles of human morality. Meanwhile, in AI ethics, the goal has been to engineer systems that can make moral decisions, in some ways inspired by how humans do this. We aim to showcase the state of the art in both fields and to show how they can be hybridized into a computational cognitive science of morality.
This virtual workshop will span a full day, split into two sessions. Each session will be composed of four talks and a discussant.
The first session, “Reverse-Engineering the Morality of Humans”, will focus on the ways that human moral judgment can be studied using a reverse-engineering approach. Talks in this segment will present research using computational methods including computational cognitive modeling, rational analysis, and game theory.
The second session, “Learning from Humans to Build Moral AI”, will showcase a series of proposals for building ethical AI that draw insights from cognitive science. Talks in this segment look to how human cognition navigates the complex moral world as a starting place to generate engineering solutions to similar problems.
A poster session will be held between the two segments, highlighting recent work from emerging scholars on either theme. Posters will be presented (along with informal socializing) on gather.town. We will end the workshop with a panel discussion between all speakers and organizers.
To attend the workshop, you will need to register for the annual meeting of the Cognitive Science Society. To submit to the workshop, see details below.
The full workshop description can be found here.
Call for Abstracts
Submissions for poster presentations are now being accepted. The poster session will be held between the morning and afternoon sessions on gather.town. In addition, 1-2 additional speakers may be selected to present talks. Contributions related to engineering or reverse-engineering morality are welcome, as are contributions from a broad range of disciplines including philosophy, computer science, psychology, legal studies, and others. We especially encourage emerging scholars and members of under-represented groups to apply.
Please submit an abstract of 250-500 words here. Deadline: June 20.
Questions about the submission process should be directed to Sydney Levine (smlevine@mit.edu).
Invited Speakers
Jean-Baptiste André
French National Centre for Scientific Research (CNRS), Department of Cognitive Sciences
Nicolas Baumard
Research Director at the CNRS and Professor at
the Ecole Normale Superieure.
Sholei Croom
Johns Hopkins, Psychological and Brain Sciences
Alison Gopnik
UC Berkeley
Psychology, Philosophy
Julia Haas
DeepMind
Ethics Research Team
Gillian Hadfield
University of Toronto
Law and Strategic Management
Dylan Hadfield-Menell
UC Berkeley/MIT
Electrical Engineering and Computer Sciences
Shaun Nichols
Cornell
Philosophy
Peter Railton
University of Michigan
Philosophy
Stuart Russell
UC Berkeley
Computer Science and Engineering
Henry Shevlin
University of Cambridge
Leverhulme Centre for the Future of Intelligence
Organizers
Sydney Levine
Harvard, Psychology
MIT, Brain and Cognitive Sciences
Fiery Cushman
Harvard, Psychology
Joshua Tenenbaum
MIT, Brain and Cognitive Sciences
Iyad Rahwan
Max Planck Institute for Human Development
Workshop Schedule
Times are Eastern (EDT)
10:00 – Jean-Baptiste André & Nicolas Baumard
10:30 – Shaun Nichols
11:00 – Sydney Levine, Josh Tenenbaum, Fiery Cushman
11:30 – short break
11:40 – Gillian Hadfiled, 20 min discussant talk + group discussion of morning sessions
12:30 – 90-second advertisements for posters
12:45 – posters on gather.town + lunch break
1:45 – Dylan Hadfield-Menell & Stuart Russell
2:15 – Julia Haas
2:45 – Alison Gopnik
3:15 – Peter Railton
3:45 – short break
3:55 – Henry Shevlin, 20 min discussant talk + group discussion of afternoon talks
4:45 – Sholei Croom facilitates general discussion with questions for participants
5:15 – opportunity for additional general discussion on any topic related to the workshop
5:45 – end
Posters
Note: Use the poster numbers to locate the posters on gather.town.
(Link available in Underline.)
Parker Crutchfield (Western Michigan University Homer Stryker M.D. School of Medicine) & Scott Scheall (Arizona State University), "Ignorance and Moral Judgment"
Milan Andrejević, Joshua White, Daniel Feuerriegel, Simon Laham, Stefan Bode (University of Melbourne, Australia), "Response time modelling reveals evidence for multiple, distinct sources of moral decision caution"
Enda Tan & J. Kiley Hamlin (University of British Columbia), "Probing the links between goal understanding and sociomoral evaluation in infancy using eye-tracking"
Léo Fitouchi, Jean-Baptiste André, Nicolas Baumard (Institut Jean Nicod, ENS, Paris), "Moral disciplining: the cognitive and evolutionary foundations of puritanical morality"
Cillian McHugh (University of Limerick), Marek McGann (Mary Immaculate College), Eric R. Igou (University of Limerick) & Elaine L. Kinsella (University of Limerick), "Moral Judgment as Categorization (MJAC)"
Rafal Rzepka, Yuki Katsumata & Kenji Araki (Hokkaido University), "Current Language Models Might Not Be Suitable For Reverse Engineering Moral Wisdom of Crowds"
Neele Engelmann & Michael R. Waldmann (Georg-August-University, Göttingen), "How to weigh lives. A computational model of moral judgment in multiple-outcomes structures"
Sarah Wu, Tobias Gerstenberg (Stanford), "The role of counterfactual reasoning in responsibility judgments"
This workshop is hosted by the 43rd Annual Meeting of the Cognitive Science Society.