Non-prioritized Belief Change (Artificial Intelligence: Foundations, Theory, and Algorithms)

Belief change (also referred to as logic of theory change or as belief revision) is a research area that models how rational agents modify their beliefs in response to new information or experiences. It emerged as a major field of study in the mid-1980s and has since become a key topic across disciplines such as philosophy, logic, and artificial intelligence. The one that is nowadays considered the standard model of belief change is the AGM model, proposed by Alchourrón, Gärdenfors and Makinson in the Journal of Symbolic Logic in 1985 (the so-called AGM paper).

In the four decades that have followed since, the area has developed rapidly and in many directions. The AGM model inspired many researchers to propose extensions and generalizations (for an overview see Belief Change by Fermé and Hansson, Springer, 2018), from among which the authors highlight the models of non-prioritized belief change: These are models in which the belief change operators considered do not give priority to the new information received (contrary to what is the case regarding the AGM model which fulfils the principle of primacy of new information). For example, the output of a non-prioritized revision may not contain the new belief that has motivated that revision. Analogously, the outcome of a non-prioritized contraction may still contain the sentence by which the contraction is made.

In a deep and comprehensive manner, this book summarises the contributions made to the modelling of non-prioritized belief change in the literature from 1985. All models are presented and taxonomized according to their main characteristics.

Eduardo Fermé is a Full Professor at the University of Madeira, where he currently heads the Department of Informatics Engineering and Interactive Media Design. He holds a PhD in Computer Science from the University of Buenos Aires (1999) and a PhD in Philosophy from KTH Royal Institute of Technology, Sweden (2011). His research focuses on the foundations of belief change, with an emphasis on bridging formal models with practical applications in Artificial Intelligence, including areas such as knowledge representation, non-monotonic logic, and decision-making systems. He has published more than 40 journal articles—over half in top-ranked international journals—and over 40 peer-reviewed conference papers, with regular contributions to major AI venues such as IJCAI, AAAI, and KR. He has led more than 10 research projects and contributed as a senior researcher to several others, funded by agencies in Portugal, Argentina, Brazil, France, Germany, and the European Union. As a lecturer, he has taught for over 30 years in the areas of Logic, Artificial Intelligence, and Belief Revision at undergraduate and postgraduate levels in Argentina, Portugal, and Brazil. He has supervised over 40 master’s dissertations, 6 doctoral theses, and 4 postdoctoral researchers in diverse fields including Informatics, Mathematics, Psychology, and Education.

Marco Garapa is a teacher of the Secretaria Regional de Educação, Ciência e Tecnologia of Madeira and has been teaching for over two decades at the Faculty of Exact Sciences and Engineering of the University of Madeira, where he carries out his academic activities. He graduated in Mathematics in 1999, completed a Master’s degree in Mathematics in 2006, and obtained a PhD in Mathematics, with a specialization in Logic and Computer Science in 2017, all from University of Madeira. He is a researcher at CIMA (Centro de Investigação em Matemática e Aplicações). His primary research interests lie in belief revision, particularly in non-prioritized belief change and belief base dynamics. His research findings have been published in leading international journals and presented at prominent conferences within the fields of Artificial Intelligence and Logic. Throughout his academic career, he has taught a broad range of courses at both undergraduate and master’s levels, including subjects such as Logic, Computational Logic, and Calculus, among others.

Maurício D. L. Reis is an Associate Professor in the Department of Mathematics at the Faculty of Exact Sciences and Engineering of the University of Madeira (UMa), and a researcher at CIMA (Centro de Investigação em Matemática e Aplicações). He obtained a PhD in Mathematics, specialising in Logic and Computer Science, from UMa in 2011. Prior to that, he graduated in Mathematics from UMa in 2001 and earned an MSc in Applied Mathematics from Instituto Superior Técnico – Technical University of Lisbon, in 2004. His main research interests are in the field of the Logic of Belief Change, and his principal scientific contributions concern multiple contraction, belief base revision, and non-prioritised belief change. His scholarly work includes several articles published in leading scientific journals and in the proceedings of top-tier scientific conferences. With over two decades of teaching experience, his pedagogical activities have encompassed a broad spectrum of courses at both undergraduate and master’s levels, covering subjects including, among others, Logic, Applied Logic, and Computability Theory.