In many real-world applications of data mining, datasets can be represented using matrices, where rows of the matrix correspond to objects (or data instances) and columns to features (or attributes). Often the datasets are in high-dimensional feature space. For example, in the vector space model of text data, the feature dimension is the vocabulary size. If representing a social network using an adjacency matrix, the feature dimension corresponds to the number of objects in the network. Many other datasets also fall into this category, such as genetic datasets, images, and medical datasets. Even though the feature dimension is enormous, a common observation is that the high-dimensional datasets may (approximately) lie in a subspace of s...
Kernel embedding of distributions has led to many recent advances in machine learning. However, late...
Dimension reduction is a research hotspot in recent years, especially manifold learning for high-dim...
Unsupervised learning involves inferring the inherent structures or patterns from unlabeled data. Si...
In many real-world applications of data mining, datasets can be represented using matrices, where ro...
We study two major topics on statistical inference for high dimensional data with low rank structure...
We study the problem of embedding high-dimensional visual data into low-dimensional vector represent...
Thesis (Ph.D.)--University of Washington, 2015The topic of learning matrix structures in the emph{hi...
Tensors and multiway analysis aim to explore the relationships between the variables used to represe...
In the real world, many kinds of high-dimensional data, such as images, documents, user-rating data,...
In many signal processing, machine learning and computer vision applications, one often has to deal ...
We present an extension of the recently introduced Generalized Matrix Learning Vector Quantization a...
Learning generative probabilistic models is a core problem in machine learning, which presents signi...
The main focus of my research is to design effective learning techniques for information retrieval a...
This thesis shows how we can exploit low-dimensional structure in high-dimensional statistics and ma...
The problem of dimensionality reduction arises in many fields of information processing, including m...
Kernel embedding of distributions has led to many recent advances in machine learning. However, late...
Dimension reduction is a research hotspot in recent years, especially manifold learning for high-dim...
Unsupervised learning involves inferring the inherent structures or patterns from unlabeled data. Si...
In many real-world applications of data mining, datasets can be represented using matrices, where ro...
We study two major topics on statistical inference for high dimensional data with low rank structure...
We study the problem of embedding high-dimensional visual data into low-dimensional vector represent...
Thesis (Ph.D.)--University of Washington, 2015The topic of learning matrix structures in the emph{hi...
Tensors and multiway analysis aim to explore the relationships between the variables used to represe...
In the real world, many kinds of high-dimensional data, such as images, documents, user-rating data,...
In many signal processing, machine learning and computer vision applications, one often has to deal ...
We present an extension of the recently introduced Generalized Matrix Learning Vector Quantization a...
Learning generative probabilistic models is a core problem in machine learning, which presents signi...
The main focus of my research is to design effective learning techniques for information retrieval a...
This thesis shows how we can exploit low-dimensional structure in high-dimensional statistics and ma...
The problem of dimensionality reduction arises in many fields of information processing, including m...
Kernel embedding of distributions has led to many recent advances in machine learning. However, late...
Dimension reduction is a research hotspot in recent years, especially manifold learning for high-dim...
Unsupervised learning involves inferring the inherent structures or patterns from unlabeled data. Si...