Temporal Calculus of Conditionals

Title: Temporal Calculus of Conditionals

Research Question: How can we embed existing conditional event algebras (CEAs) into a three-valued extension of temporal logic of discrete past time, known as (TL|TL), to improve their descriptive power and computational efficiency?

Methodology: The authors propose a framework that combines the tools of (TL|TL), Moore machines, Markov chains, and conditional objects to embed existing CEAs. They introduce present and past tense connectives to represent conditional events and their relationships.

Results: The authors demonstrate that the embeddings allow for the use of native (TL|TL) algorithms for computing probabilities of complex conditional expressions, which can outperform previous methods. They also show that the embeddings address the descriptive incompleteness of existing CEAs, allowing for a more precise formulation and analysis of important notions like independence of conditional events.

Implications: The research has significant implications for the field of probabilistic reasoning, particularly in domains such as Bayesian methods, knowledge discovery in databases, and decision sciences. The proposed framework provides a more powerful and computationally efficient model for representing and reasoning with conditional events, which can lead to improved performance in various applications.

Link to Article: https://arxiv.org/abs/0110004v1 Authors: arXiv ID: 0110004v1