Abstract:Spectral methods have recently emerged as a powerful tool for dimensionality reduction and manifold learning. These methods use information contained in the eigenvectors of a data affinity (i.e., item-item similarity) matrix to reveal the low dimensional structure in the high dimensional data. The most popular manifold learning algorithms include Locally Linear Embedding, ISOMAP, and Laplacian Eigenmap. However, these algorithms only provide the embedding results of training samples. There are many extensions of these approaches which try to solve the out-of-sample extension problem by seeking an embedding function in reproducing kernel Hilbert space. However, a disadvantage of all these approaches is that their computations usually involve eigen-decomposition of dense matrices which is expensive in both time and memory. In this thesis, we introduce a novel dimensionality reduction framework, called Spectral Regression (SR). SR casts the problem of learning an embedding function into a regression framework, which avoids eigen-decomposition of dense matrices. Also, with the regression as a building block, different kinds of regularizers can be naturally incorporated into our framework which makes it more flexible. SR can be performed in supervised, unsupervised and semi-supervised situation. It can make efficient use of both labeled and unlabeled points to discover the intrinsic discriminant structure in the data. We have applied our algorithms to several real world applications, e.g. face analysis, document representation and content-based image retrieval.

Spectral Partial Least Squares Regression

Regularized direct Linear Graph Embedding

Unsupervised Large Graph Embedding

Spectral regression: a regression framework for efficient regularized subspace learning

Extreme spectral regression for efficient regularized subspace learning.

Learning spectral embedding via iterative eigenvalue thresholding.

Graph-regularized Feature Selection Based on Spectral Learning and Subspace Learning

Spectral Regression: A Unified Approach for Sparse Subspace Learning

Low Rank Spectral Regression Via Matrix Factorization for Efficient Subspace Learning

Regularized Locality Preserving Indexing Via Spectral Regression

Sparse Representation Based Spectral Regression

Joint Low-Rank Representation and Spectral Regression for Robust Subspace Learning.

Spectral Regression for Efficient Regularized Subspace Learning

Discriminant Graph Based Linear Embedding.

GRASPEL: Graph Spectral Learning at Scale

Spatial-Spectral Graph-Based Nonlinear Embedding Dimensionality Reduction for Hyperspectral Image Classificaiton.

Spectral clustering with linear embedding: A discrete clustering method for large-scale data

Spectral embedding of graphs

Unsupervised Large Graph Embedding Based on Balanced and Hierarchical K-Means.

A framework of joint graph embedding and sparse regression for dimensionality reduction.

Low-Rank Projection Learning Via Graph Embedding