Contexts: a formalization and some applications
Contexts: a formalization and some applications
Computation as logic
Multilanguage hierarchical logics, or: how we can do without modal logics
Artificial Intelligence
A vademecum of ambivalent logic
Meta-logics and logic programming
Hierarchical Meta-Logics: Intuitions, Proof Theory and Semantics
META-92 Proceedings of the 3rd International Workshop on Meta-Programming in Logic
IJCAI'93 Proceedings of the 13th international joint conference on Artifical intelligence - Volume 1
Quantificational logic of context
AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1
Contextual reasoning is NP-complete
AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1
Fundamenta Informaticae
Fundamenta Informaticae
Agent-Oriented Language Engineering for Robust NLP
ESAW '01 Proceedings of the Second International Workshop on Engineering Societies in the Agents World II
Improving the Efficiency of Reasoning Through Structure-Based Reformulation
SARA '02 Proceedings of the 4th International Symposium on Abstraction, Reformulation, and Approximation
First-Order Contextual Reasoning
SBIA '02 Proceedings of the 16th Brazilian Symposium on Artificial Intelligence: Advances in Artificial Intelligence
Partition-based logical reasoning for first-order and propositional theories
Artificial Intelligence - Special volume on reformulation
Partition-based logical reasoning for first-order and propositional theories
Artificial Intelligence - Special volume on reformulation
Hi-index | 0.00 |
We consider the problem of building an automated proof system for reasoning in contexts. Towards that goal, we first define a language of contextual implications, and give its operational semantics under the form of a natural deduction system using explicit context assertions. We show that this proof system has an equivalent straightforward logic program, which in tum can be reified, i.e. defined as an outer meta-level context, and thus applied to itself. More powerful reasoning models (e.g. those involving theory lifting) can be then implemented by applying the same logic program on extended meta-level contexts containing specialized axioms. As a theoretical application, we consider the task of concept learning. In order to achieve generality (Le. abstracting solution classes from problem instances), we argllle that concept learning goals should aim at the discovery of meta-level operators representing the sequence of inference steps leading to object-level moves or actions. We illustrate: this idea with the definition of a learning model based on partial deduction with respect to theory lifting.