AI alignment
- Submission status
- Open
- Submission deadline
The process of aligning Artificial Intelligence (AI) according to human and societal values, commonly referred to as AI alignment, constitutes a fundamental quest of AI research. This highly interdisciplinary field, which encompasses computer science, ethics, psychology, and engineering, has been conceived to ensure that novel AI systems are designed and deployed in a way that maximizes their beneficial integration into society, while minimizing their risks, such as threats to human safety or autonomy. This includes ensuring that these systems are designed to respect human values, are transparent and interpretable, and can reason about their own behaviour.
AI alignment problem is difficult both to approach and to solve, for several reasons; one of the main challenges is the complexity of human values and preferences. In the end, at the centre of the field lay human values, which not only constitute inherently diverse and complex sets of societal characteristics, but also vary over time, making their defining and formalisation a highly demanding encoding task.
Other major challenges in the field come from ethical and safety considerations, ranging from bias in algorithm training datasets, to interpretation in decision-making. Finally, as AI developments grow rapidly, and span both theoretical, computational subfields of AI research, as well as highly applied sciences, such as mechanical and electrical engineering, the field faces a truly unique set of challenges that need to be addressed in real time.
This Collection is dedicated to the latest developments in the heterogenous field of AI alignment. As the field relies on a range of technologies and tools for its growth, from cyber-physical systems, to privacy, to ubiquitous computing, we welcome previously unpublished research in computer science, machine learning, and engineering, which carries the potential to address some of the many challenges this exciting field is presently dealing with. Please provide a cover letter with your submission in which you clearly describe the ways in which your manuscript addresses the problem of AI alignment.
Editors
-
Ayman El-Baz
University of Louisville, USA
-
Jose M. Molina
Universidad Carlos III de Madrid, Spain
-
Lianyong Qi
China University of Petroleum (East China), China
-
Michael A. Riegler
Simula Metropolitan Center for Digital Engineering & Oslo Metropolitan University, Norway
