Learning engineering models with the minimum description length principle

Authors:
R. Bharat Rao;Stephen C.-Y. Lu
Affiliations:
Knowledge-Based Engineering Systems Laboratory, University of illinois at Urbana-Champaign;Knowledge-Based Engineering Systems Laboratory, University of illinois at Urbana-Champaign
Venue:
AAAI'92 Proceedings of the tenth national conference on Artificial intelligence
Year:
1992

Citing 9
Cited 1

Scientific discovery: computational explorations of the creative process

Scientific discovery: computational explorations of the creative process
Inferring decision trees using the minimum description length principle

Information and Computation
Part segmentation for object recognition

Neural Computation
Conceptual set covering: improving fit-and-split algorithms

Proceedings of the seventh international conference (1990) on Machine learning
Information Theory and Reliable Communication

Information Theory and Reliable Communication
Induction of Decision Trees

Machine Learning
Integrating Quantitative and Qualitative Discovery: The ABACUS System

Machine Learning
Conjunctive conceptual clustering: a methodology and experimentation (learning)

Conjunctive conceptual clustering: a methodology and experimentation (learning)
Some experiments in applying inductive inference principles to surface reconstruction

IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence - Volume 2

Automated Discovery Of Empirical Laws

Fundamenta Informaticae

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper discusses discovery of mathematical models from engineering data sets. KEDS, a Knowledge-based Equation Discovery System, identifies several potentially overlapping regions in the problem space, each associated with an equation of different complexity and accuracy. The minimum description length principle, together with the KEDS algorithm, is used to guide the partitioning of the problem space. The KEDSMDL algorithm has been tested on discovering models for predicting the performance efficiencies of an internal combustion engine.