Forming concepts for fast inference

Authors:
Henry Kautz;Bart Selman
Affiliations:
AI Principles Research Department, AT&T Bell Laboratories, Murray Hill, NJ;AI Principles Research Department, AT&T Bell Laboratories, Murray Hill, NJ
Venue:
AAAI'92 Proceedings of the tenth national conference on Artificial intelligence
Year:
1992

Citing 8
Cited 5

Boolean Feature Discovery in Empirical Learning

Machine Learning
The complexity of finite functions

Handbook of theoretical computer science (vol. A)
Computational Complexity of Machine Learning

Computational Complexity of Machine Learning
A General Framework for Knowledge Compilation

PDK '91 Proceedings of the International Workshop on Processing Declarative Knowledge
The complexity of theorem-proving procedures

STOC '71 Proceedings of the third annual ACM symposium on Theory of computing
Learning effective search control knowledge: an explanation-based approach

Learning effective search control knowledge: an explanation-based approach
The relevance of irrelevance

IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 1
Selectively generalizing plans for problem-solving

IJCAI'85 Proceedings of the 9th international joint conference on Artificial intelligence - Volume 1

First order LUB approximations: characterization and algorithms

Artificial Intelligence - Special volume on reformulation
In defense of PDDL axioms

Artificial Intelligence
Representing CSPs with set-labeled diagrams: a compilation map

GKR'11 Proceedings of the Second international conference on Graph Structures for Knowledge Representation and Reasoning
Computation of Extensions of Seminormal Default Theories

Fundamenta Informaticae
Approximate Inference In Default Logic And Circumscription

Fundamenta Informaticae

Quantified Score

Hi-index	0.00

Visualization

Abstract

Knowledge compilation speeds inference by creating tractable approximations of a knowledge base, but this advantage is lost if the approximations are too large. We show how learning concept generalizations can allow for a more compact representation of the tractable theory. We also give a general induction rule for generating such concept generalizations. Finally, we prove that unless NP ⊆ non-uniform P, not all theories have small Horn least-upper-bound approximations.