Explanation-Augmented SVM: an approach to incorporating domain knowledge into SVM learning

Authors:
Qiang Sun;Gerald DeJong
Affiliations:
University of Illinois at Urbana-Champaign, Urbana, IL;University of Illinois at Urbana-Champaign, Urbana, IL
Venue:
ICML '05 Proceedings of the 22nd international conference on Machine learning
Year:
2005

Citing 8
Cited 2

Generalization performance of support vector machines and other pattern classifiers

Advances in kernel methods
Machine Learning

Machine Learning
Training Invariant Support Vector Machines

Machine Learning
Latent Semantic Kernels

Journal of Intelligent Information Systems
Artificial Intelligence: A Modern Approach

Artificial Intelligence: A Modern Approach
Feature Kernel Functions: Improving SVMs Using High-Level Knowledge

CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
Structural risk minimization over data-dependent hierarchies

IEEE Transactions on Information Theory
On the generalization of soft margin algorithms

IEEE Transactions on Information Theory

Explanation-based learning for image understanding

AAAI'06 proceedings of the 21st national conference on Artificial intelligence - Volume 2
Generative prior knowledge for discriminative classification

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

We introduce a novel approach to incorporating domain knowledge into Support Vector Machines to improve their example efficiency. Domain knowledge is used in an Explanation Based Learning fashion to build justifications or explanations for why the training examples are assigned their given class labels. Explanations bias the large margin classifier through the interaction of training examples and domain knowledge. We develop a new learning algorithm for this Explanation-Augmented SVM (EA-SVM). It naturally extends to imperfect knowledge, a stumbling block to conventional EBL. Experimental results confirm desirable properties predicted by the analysis and demonstrate the approach on three domains.