The 2nd Computational Machine Ethics (CME) Workshop will be held in conjunction with the 20th International Conference on Principles of Knowledge Representation and Reasoning (KR 2023). It will provide a forum for discussing recent research on methodological development in machine ethics and its impact on human societies. With the successful development of cognitive machines in a diverse range of application scenarios, there is an urgent need to enable these machines to behave morally following the behavioural norms and ethical principles established in human societies. Machine ethics is thus a subfield of AI ethics concerned with implementing moral and ethical behaviours in cognitive machines, i.e., to develop computational approaches to ensure ethical consideration in AI applications. In this way, cognitive machines will be adequately equipped with the moral and ethical competence to prevent physical, emotional or psychological harm and to maximise the benefits they can provide to society. In recent years various approaches have been developed, which can generally be categorised into logic-based formulations, learning-based methods and hybrid models. Nevertheless, compared to the extensive discussion in the press regarding the dangers of AI and rapid development of frameworks for regulating the ethical use of AI, machine ethics remains a relatively less explored field. In particular, how to represent ethical principles and reason about the best course of action for real-life scenarios in a generalisable way largely remains an open question. 

Continuing from the 1st CME Workshop that was held in conjunction with KR 2021, the 2nd CME workshop aims to provide a forum for researchers from academia and industry to exchange ideas and techniques in the area of machine ethics, to generate new ideas in the intersection of knowledge representation and reasoning and machine/deep learning, to promote widespread awareness of machine ethics in the AI community as well as identifying new research directions in the area. Besides regular research papers, we also welcome vision papers, demonstration papers and papers showcasing industry applications.

Topics of interest include but are not limited to:

• Computational modelling of morality and ethics

• Explicit and implicit ethical agents

• Logic-based approaches to machine ethics

• Knowledge representation of ethical principles

• Machine learning-based approaches to machine ethics

• Natural language processing for descriptive ethics

• Integration of symbolic and sub-symbolic models

• Commonsense reasoning and its application to machine ethics

• Explainability of ethical reasoning

• Formal verification of machine ethics

• Development of machine ethics in cognitive robot programs

• Robot learning for ethical reasoning

3 Sep 2023, 8:30am - 12:45pm (Tentative schedule)

• 8:30-9am: Arrival, Networking and Opening

• 9-9:50am: Keynote - Professor Marija Slavkovik

• 9:50-10:10am: Paper presentation 1

  • • • • Yousef Taheri, Gauvain Bourgne and Jean-Gabriel Ganascia

        • Modelling Integration of Responsible AI Values for Ethical Decision Making

• 10:10-10:30am: Paper presentation 2

  • • • • Simon Kolker, Louise Dennis, Ramon Fraga Pereira and Mengwei Xu

        • Selecting Ethical Actions by Retrospection on Hypothetical Outcomes

• 10:30-11am: Coffee Break

• 11-11:20am: Paper presentation 3

  • • • • Adam Wyner, Tomasz Zurek and Dorota Stachura-Zurek

        • Towards a Formalisation of Value-based Actions and Consequentialist Ethics

• 11:20-11:40am: Paper presentation 4

  • • • • Paulus Meessen

        • Investigations on how Computer Scientists believe the Law should work

• 11:40am-12pm: Invited talk 1 - A/Prof Andrea Loreggia

  • • • • Making Human-Like Moral Decisions 

• 12-12:10pm: Invited talk 2 - Dr Valentina Pyatkin

  • • • • ClarifyDelphi: Reinforced Clarification Questions with Defeasibility Rewards for Social and Moral Situations

• 12:10-12:30pm: Invited talk 3 - Taylor Olson

  • • • • Mitigating Adversarial Norm Training with Moral Axioms 

    • 12:30-12:45pm: Closing

Professor Marija Slavkovik, University of Bergen, Norway

Machine Ethics: from SF to practice

Machine ethics, as defined in the 2006 AAAI symposium is "concerned with the behaviour of machines towards human users and other machines”. The field spans philosophy and computer science and it is studied from both disciplines. It is also the filed in which we get to cite SF work as motivation. Machine ethics forces us to revisit hard AI problems of emulating behaviour which is difficult to formally specify. When we think of the practical implementations of machine ethics what comes to mind are scenarios of robots helping the elderly or driverless vehicles. Is how we think of AI biasing our view on what relevant practical applications of machine ethics are? I give an overview of the machine ethics field from SF to main breakthroughs,  propose that there are more interesting problems of machine ethics in content moderation than on the road, and discuss what in my view are the delicious challenges we can address as AI researchers. 

Marija Slavkovik is a Professor with the Faculty for Social Sciences of the University of Bergen. Her background is in computer science and artificial intelligence.  Marija works on formalising ethical collective decision-making. She has held held several seminars, tutorials and graduate courses on AI ethics. Marija is a vice-chair of the the Norwegian AI Association,  board member of European Association for Artificial Intelligence, a member of the informal advisory group on Ethics, Legal, Social Issues (ELS) of CLAIRE, and AI and Society track editor of JAIR. In her free time she writes project proposals and reads SF. 

Maurice Pagnucco, University of New South Wales, Australia

Yang Song, University of New South Wales, Australia

Louise Dennis, University of Manchester, UK

Michael Anderson, University of Hartford, USA

Ilaria Canavotto, University of Maryland, USA

Jean-Gabriel Ganascia, Sorbonne University, France 

Randy Goebel, University of Alberta, Canada 

Aleks Knoks, University of Luxembourg, Luxembourg

Gerhard Lakemeyer, RWTH Aachen, Germany 

Beishui Liao, Zhejiang University, China 

Emiliano Lorini, University Paul Sabatier, France

Katina Michael, Arizona State University, USA 

Vivek Nallur, University College Dublin, Ireland

Abhaya Nayak, Macquarie University, Australia 

Ken Satoh, National Institute of Informatics, Japan 

Anthony Tompkins, University of New South Wales, Australia

Leon van der Torre, University of Luxembourg, Luxembourg

Paper submission: 11 June 2023

Paper notification: 4 July 2023

Workshop date and time: 3 Sep 2023, 8:30am - 12:45pm

We welcome and encourage the submission of original, previously unpublished research and also have a recently published research track, similarly to the main conference of KR 2023. Submissions should be 3-6 pages (including figures, references, etc.) and formatted according to the KR 2023 guidelines. Submission is now open in EasyChair: https://easychair.org/my/conference?conf=kr2023, as a special track under KR 2023.