Weighted Maximum Variance Dimensionality Reduction
نویسندگان
چکیده
Dimensionality reduction procedures such as principal component analysis and the maximum margin criterion discriminant are special cases of a weighted maximum variance (WMV) approach. We present a simple two parameter version of WMV that we call 2P-WMV. We study the classification error given by the 1-nearest neighbor algorithm on features extracted by our and other dimensionality reduction methods on several real datasets. Our results show that our method yields the lowest average error across the datasets with statistical significance.
منابع مشابه
On a Connection between Maximum Variance Unfolding, Shortest Path Problems and IsoMap
We present an equivalent formulation of the Maximum Variance Unfolding (MVU) approach to nonlinear dimensionality reduction in terms of distance matrices. This yields a novel interpretation of the MVU problem as a regularized version of the shortest path problem on a graph. This interpretation enables us to establish an asymptotic convergence result for the case that the underlying data are dra...
متن کاملAn Introduction to Nonlinear Dimensionality Reduction by Maximum Variance Unfolding
Many problems in AI are simplified by clever representations of sensory or symbolic input. How to discover such representations automatically, from large amounts of unlabeled data, remains a fundamental challenge. The goal of statistical methods for dimensionality reduction is to detect and discover low dimensional structure in high dimensional data. In this paper, we review a recently proposed...
متن کاملColored Maximum Variance Unfolding
Maximum variance unfolding (MVU) is an effective heuristic for dimensionality reduction. It produces a low-dimensional representation of the data by maximizing the variance of their embeddings while preserving the local distances of the original data. We show that MVU also optimizes a statistical dependence measure which aims to retain the identity of individual observations under the distancep...
متن کاملSpectral Dimensionality Reduction via Maximum Entropy
We introduce a new perspective on spectral dimensionality reduction which views these methods as Gaussian random fields (GRFs). Our unifying perspective is based on the maximum entropy principle which is in turn inspired by maximum variance unfolding. The resulting probabilistic models are based on GRFs. The resulting model is a nonlinear generalization of principal component analysis. We show ...
متن کاملA Unifying Probabilistic Perspective for Spectral Dimensionality Reduction: Insights and New Models
We introduce a new perspective on spectral dimensionality reduction which views these methods as Gaussian Markov random fields (GRFs). Our unifying perspective is based on the maximum entropy principle which is in turn inspired by maximum variance unfolding. The resulting model, which we call maximum entropy unfolding (MEU) is a nonlinear generalization of principal component analysis. We relat...
متن کامل