Unsupervised Slow Subspace-Learning from Stationary Processes

Abstract. We propose a method of unsupervised learning from stationary, vector-valued processes. A low-dimensional subspace is selected on the basis of a criterion which rewards data-variance (like PSA) and penalizes the variance of the velocity vector, thus exploiting the shorttime dependencies of the process. We prove error bounds in terms of the -mixing coe¢ cients and consistency for absolutely regular processes. Experiments with image recognition demonstrate the algorithms ability to learn geometrically invariant feature maps.

• Type: Conference
• Year: 2006
• Where: ALT
• Authors: Andreas Maurer
• Keywords: ALT 2006, Error Bounds, Invariant Feature Maps, Low-dimensional Subspace, Machine Learning