Information Processing Letters
Self bounding learning algorithms
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
Generalization in decision trees and DNF: does size matter?
NIPS '97 Proceedings of the 1997 conference on Advances in neural information processing systems 10
Microchoice bounds and self bounding learning algorithms
COLT '99 Proceedings of the twelfth annual conference on Computational learning theory
Learning in Neural Networks: Theoretical Foundations
Learning in Neural Networks: Theoretical Foundations
A Fast, Bottom-Up Decision Tree Pruning Algorithm with Near-Optimal Generalization
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Generalization Bounds for Decision Trees
COLT '00 Proceedings of the Thirteenth Annual Conference on Computational Learning Theory
Rademacher and gaussian complexities: risk bounds and structural results
The Journal of Machine Learning Research
Tutorial on Practical Prediction Theory for Classification
The Journal of Machine Learning Research
Estimation of Dependences Based on Empirical Data: Springer Series in Statistics (Springer Series in Statistics)
Data Mining: Practical Machine Learning Tools and Techniques, Second Edition (Morgan Kaufmann Series in Data Management Systems)
Process-Specific Information for Learning Electronic Negotiation Outcomes
Fundamenta Informaticae
Sample compression, margins and generalization: extensions to the set covering machine
Sample compression, margins and generalization: extensions to the set covering machine
Hierarchical linear support vector machine
Pattern Recognition
Hi-index | 0.00 |
We propose a formulation of the Decision Tree learning algorithm in the Compression settings and derive tight generalization error bounds. In particular, we propose Sample Compression and Occam's Razor bounds. We show how such bounds, unlike the VC dimension or Rademacher complexities based bounds, are more general and can also perform a margin-sparsity trade-off to obtain better classifers. Potentially, these risk bounds can also guide the model selection process and replace traditional pruning strategies.