Building robust learning systems by combining induction and optimization

Authors:
David Tcheng;Bruce Lambert;Stephen C-Y. Lu;Larry Rendell
Affiliations:
Computer Science, University of Illinois, Urbana, IL;Speech Communication, University of Illinois, Urbana, IL;Mechanical Engineering, University of Illinois, Urbana, IL;Computer Science, University of Illinois, Urbana, IL
Venue:
IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence - Volume 1
Year:
1989

Citing 6
Cited 3

Machine learning of inductive bias

Machine learning of inductive bias
Generalized recursive splitting algorithms for learning hybrid concepts

Proceedings of the sixth international workshop on Machine learning
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations
Adaptation in natural and artificial systems

Adaptation in natural and artificial systems
Induction of Decision Trees

Machine Learning
Human Problem Solving

Human Problem Solving

How to shift bias: Lessons from the baldwin effect

Evolutionary Computation
Myths and legends in learning classification rules

AAAI'90 Proceedings of the eighth National conference on Artificial intelligence - Volume 2
Inductive policy

AAAI'92 Proceedings of the tenth national conference on Artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Each concept description language and search strategy has an inherent inductive bias, a preference for some hypotheses over others. No single inductive bias performs optimally on all problems. This paper describes a system that couples induction with optimization to carry out an efficient search of large regions of inductive bias space. Experimental results are reported demonstrating the system's capacity to choose optimal biases even for complex and noisy problems.