[NIPS 2012] Projection Retrieval for Classification英文文献资料.pdf

Projection Retrieval for Classification Madalina Fiterau Artur Dubrawski Machine Learning Department School of Computer Science Carnegie Mellon University Carnegie Mellon University Pittsburgh, PA 15213 Pittsburgh, PA 15213 mfiterau@cs.cmu.edu awd@cs.cmu.edu Abstract In many applications, classification systems often require human intervention in the loop. In such cases the decision process must be transparent and comprehen- sible, simultaneously requiring minimal assumptions on the underlying data dis- tributions. To tackle this problem, we formulate an axis-aligned subspace-finding task under the assumption that query specific information dictates the comple- mentary use of the subspaces. We develop a regression-based approach called RECIP that efficiently solves this problem by finding projections that minimize a nonparametric conditional entropy estimator. Experiments show that the method is accurate in identifying the informative projections of the dataset, picking the correct views to classify query points, and facilitates visual evaluation by users. 1 Introduction and problem statement In the domain of predictive analytics, many applications which keep human users in the loop require the use of simple classification models. Often, it is required that a test-point be ‘explained’ (classi- fied) using a simple low-dimensional projection of the original feature space. This is a Projection Retrieval for Classification problem (PRC). The interaction with the user proceeds as follows: the user provides the system a query point; the system searches for a projecti