|Title||Latent Supervised Learning.|
|Publication Type||Journal Article|
|Year of Publication||2013|
|Authors||Wei, Susan, and Michael R. Kosorok|
|Journal||J Am Stat Assoc|
|Date Published||2013 Jul 01|
A new machine learning task is introduced, called latent supervised learning, where the goal is to learn a binary classifier from training labels which serve as surrogates for the unobserved class labels. A specific model is investigated where the surrogate variable arises from a two-component Gaussian mixture with unknown means and variances, and the component membership is determined by a hyperplane in the covariate space. The estimation of the separating hyperplane and the Gaussian mixture parameters forms what shall be referred to as the change-line classification problem. A data-driven sieve maximum likelihood estimator for the hyperplane is proposed, which in turn can be used to estimate the parameters of the Gaussian mixture. The estimator is shown to be consistent. Simulations as well as empirical data show the estimator has high classification accuracy.
|Alternate Journal||J Am Stat Assoc|
|Original Publication||Latent supervised learning.|
|PubMed Central ID||PMC3848255|
|Grant List||P01 CA142538 / CA / NCI NIH HHS / United States |
T32 GM067553 / GM / NIGMS NIH HHS / United States
Latent Supervised Learning.