Bias-Variance Tradeoff in Hybrid Generative-Discriminative Models

Authors:
Guillaume Bouchard
Affiliations:
-
Venue:
ICMLA '07 Proceedings of the Sixth International Conference on Machine Learning and Applications
Year:
2007

Citing 0
Cited 6

On the generative-discriminative tradeoff approach: Interpretation, asymptotic efficiency and classification performance

Computational Statistics & Data Analysis
Exponential family hybrid semi-supervised learning

IJCAI'09 Proceedings of the 21st international jont conference on Artifical intelligence
A robust semi-supervised classification method for transfer learning

CIKM '10 Proceedings of the 19th ACM international conference on Information and knowledge management
Iterative refinement of HMM and HCRF for sequence classification

PSL'11 Proceedings of the First IAPR TC3 conference on Partially Supervised Learning
A geometric view of conjugate priors

IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Three
Joint semi-supervised learning of Hidden Conditional Random Fields and Hidden Markov Models

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

Given any generative classifier based on an inexact density model, we can define a discriminative counterpart that reduces its asymptotic error rate, while increasing the estimation variance. An optimal bias-variance balance might be found using Hybrid Generative-Discriminative (HGD) approaches. In these paper, these methods are defined in a unified framework. This allow us to find sufficient conditions under which an improvement in generalization performances is guaranteed. Numerical experiments illustrate the well fondness of our statements.