|Affiliations:||Mathematical Institute of the Serbian Academy of Sciences and Arts||Title:||Towards Probabilistic Reasoning in Type Theory - The Intersection Type Case||Journal:||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)||Volume:||12012 LNCS||First page:||122||Last page:||139||Conference:||11th International Symposium on Foundations of Information and Knowledge Systems, FoIKS 2020; Dortmund; Germany; 17 February 2020 through 21 February 2020||Issue Date:||1-Jan-2020||Rank:||M33||ISBN:||978-3-030-39950*4||ISSN:||0302-9743||DOI:||10.1007/978-3-030-39951-1_8||Abstract:||
The development of different probabilistic models of uncertainty has been inspired by the rapid progress in various fields, e.g. in AI, probabilistic programming, etc. Lambda calculus is a universal model of computation suitable to express programming languages concepts. Hence, different methods for probabilistic reasoning in lambda calculus have been investigated. In this paper, we develop a formal model for probabilistic reasoning about lambda terms with intersection types, which is a combination of lambda calculus and probabilistic logic. The language of lambda calculus with intersection types is endowed with a probabilistic operator. We propose a semantics based on the possible world approach. An infinitary axiomatization is given for this system and it is proved to be sound with respect to the proposed semantics.
|Keywords:||Intersection types | Kripke-style semantics | Lambda calculus | Probabilistic reasoning | Soundness||Publisher:||Springer Link||Project:||Representations of logical structures and formal languages and their application in computing
Development of new information and communication technologies, based on advanced mathematical methods, with applications in medicine, telecommunications, power systems, protection of national heritage and education
Swiss National Science Foundation, Grant 200021 165549
Show full item record
checked on May 30, 2023
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.