Skip to content
conftrace_
2008 JMLR JMLR 2008

An Error Bound Based on a Worst Likely Assignment

Abstract

This paper introduces a new PAC transductive error bound for classification. The method uses information from the training examples and inputs of working examples to develop a set of likely assignments to outputs of the working examples. A likely assignment with maximum error determines the bound. The method is very effective for small data sets. [abs] [ pdf ][ bib ] © JMLR 2008. (edit, beta)

🐣 Hot Topic Early Bird - pac learning

Authors

Eric Bax, Augusto Callejas

Topics

Machine Learning > Core Methods > Classification Machine Learning > Optimization & Theory > Statistical Learning

Keywords

transductive learning pac learning classification error error bound
Download PDF

More from JMLR 2008

All JMLR 2008 papers →
Multi-class Discriminant Kernel Learning via Convex Programming
Bayesian Inference and Optimal Design for the Sparse Linear Model
Finite-Time Bounds for Fitted Value Iteration
Graphical Methods for Efficient Likelihood Inference in Gaussian Covariance Models
Bouligand Derivatives and Robustness of Support Vector Machines for Regression
Accelerated Neural Evolution through Cooperatively Coevolved Synapses