Modern unsupervised learning tool for analyzing general numerical data.
Robust Non-negative matrix factorization (rNMF) is a modern method for obtaining low dimensional structure from high dimensional data. It allows both automatic and semi-supervised control to handle different types of corruptions.
Estimation of parameters in 3-segment (i.e. 2 change-point) regression models with heteroscedastic variances based on both likelihood and hybrid Bayesian approaches, with and without continuity constraints at the change points.
The IMAP identifies which 5% (the target error rate) of the declared positives are likely be true false positives, based on repeated images or only single images.
SSD calculates the sample size needed to detect the differences between two sets of unordered categorical data.