Random projection ensemble classification
Richard Samworth (University of Cambridge)
Friday 17th April, 2015 15:30-16:30 Maths 516
We propose a new method for high-dimensional classification, based on careful aggregation of the results of applying a base classifier to random projections of the feature vectors into a lower-dimensional space. More precisely, the random projections are divided into non-overlapping blocks, and within each block we select the projection yielding the smallest estimate of the test error. Our random projection ensemble classifier then aggregates the results of applying the base classifier on the selected projections, with a data-driven voting threshold to determine the final assignment. We provide theoretical understanding to justify the methodology, and a simulation comparison with several other popular high-dimensional classifiers reveals its excellent finite-sample performance.