Experiments with Random Projection

Recent theoretical work has identified random projection as a promising dimensionality reduction technique for learning mixtures of Gaussians. Here we summarize these results and illustrate them by a wide variety of experiments on synthetic and real data.

• Type: Conference
• Year: 2000
• Where: UAI
• Authors: Sanjoy Dasgupta
• Keywords: Promising Dimensionality Reduction, Random Projection, UAI 2000, UAI 2008, Wide Variety