Defaults and Infinitesimals: Defeasible Inference by Nonarchimedean Entropy-Maximization

2013/02/20 by Emil Weydert, Weydert, Emil
Arts and Humanities · Computer Science · #Artificial Intelligence (cs.AI) #Bayesian Modeling and Causal Inference #FOS: Computer and information sciences #Logic, Reasoning, and Knowledge #Philosophy and History of Science #Rough Sets and Fuzzy Logic #cs.AI

paper · pdf · doi:10.48550/arxiv.1302.4988

Appears in Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence (UAI1995)

arxiv created 2013/02/20 · openalex publication_date 2013/02/20 · arxiv updated 2013/02/21 · openalex created_date 2019/07/30 · openalex updated_date 2026/07/28

Abstract

We develop a new semantics for defeasible inference based on extended probability measures allowed to take infinitesimal values, on the interpretation of defaults as generalized conditional probability constraints and on a preferred-model implementation of entropy maximization.

Related

• Unification of Maximum Entropy and Bayesian Inference via Plausible Reasoning
• Non-Monotonicity in Probabilistic Reasoning
• A Logic for Default Reasoning About Probabilities
• The relation of bias with risk in empirically constrained inferences
• Limitations of Skeptical Default Reasoning