Pinar Donmez

JMLR Journal 2011 Journal Article

Unsupervised Supervised Learning II: Margin-Based Classification Without Labels

• Krishnakumar Balasubramanian
• Pinar Donmez
• Guy Lebanon

Many popular linear classifiers, such as logistic regression, boosting, or SVM, are trained by optimizing a margin-based risk function. Traditionally, these risk functions are computed based on a labeled data set. We develop a novel technique for estimating such risks using only unlabeled data and the marginal label distribution. We prove that the proposed risk estimator is consistent on high-dimensional data sets and demonstrate it on synthetic and real-world data. In particular, we show how the estimate is used for evaluating classifiers in transfer learning, and for training classifiers with no labeled data whatsoever. [abs] [ pdf ][ bib ] &copy JMLR 2011. ( edit, beta )

JMLR Journal 2010 Journal Article

Unsupervised Supervised Learning I: Estimating Classification and Regression Errors without Labels

• Pinar Donmez
• Guy Lebanon
• Krishnakumar Balasubramanian

Estimating the error rates of classifiers or regression models is a fundamental task in machine learning which has thus far been studied exclusively using supervised learning techniques. We propose a novel unsupervised framework for estimating these error rates using only unlabeled data and mild assumptions. We prove consistency results for the framework and demonstrate its practical applicability on both synthetic and real world data. [abs] [ pdf ][ bib ] &copy JMLR 2010. ( edit, beta )

ICML Conference 2008 Conference Paper

Optimizing estimated loss reduction for active sampling in rank learning

• Pinar Donmez
• Jaime G. Carbonell