arXiv subjects

Heng Huang

Publications and source records attributed to Heng Huang.

260 records · Page 15Linked to original sources

arXiv2016

Clinical Information Extraction via Convolutional Neural Network

We report an implementation of a clinical information extraction tool that leverages deep neural network to annotate event spans and their attributes from raw clinical notes and pathology reports. Our approach uses context words and their part-of-speech tags and shape information as features. Then we hire temporal (1D) convolutional neural network to learn hidden feature representations. Finally, we use Multilayer Perceptron (MLP) to predict event spans. The empirical evaluation demonstrates that our approach significantly outperforms baselines.

cs.LG↗

arXiv2016

Enhancing Sentence Relation Modeling with Auxiliary Character-level Embedding

Neural network based approaches for sentence relation modeling automatically generate hidden matching features from raw sentence pairs. However, the quality of matching feature representation may not be satisfied due to complex semantic relations such as entailment or contradiction. To address this challenge, we propose a new deep neural network architecture that jointly leverage pre-trained word embedding and auxiliary character embedding to learn sentence meanings. The two kinds of word sequence representations as inputs into multi-layer bidirectional LSTM to learn enhanced sentence representation. After that, we construct matching features followed by another temporal CNN to learn high-level hidden matching feature representations. Experimental results demonstrate that our approach consistently outperforms the existing methods on standard evaluation datasets.

cs.CL↗

arXiv2016

Non-Greedy L21-Norm Maximization for Principal Component Analysis

Principal Component Analysis (PCA) is one of the most important unsupervised methods to handle high-dimensional data. However, due to the high computational complexity of its eigen decomposition solution, it hard to apply PCA to the large-scale data with high dimensionality. Meanwhile, the squared L2-norm based objective makes it sensitive to data outliers. In recent research, the L1-norm maximization based PCA method was proposed for efficient computation and being robust to outliers. However, this work used a greedy strategy to solve the eigen vectors. Moreover, the L1-norm maximization based objective may not be the correct robust PCA formulation, because it loses the theoretical connection to the minimization of data reconstruction error, which is one of the most important intuitions and goals of PCA. In this paper, we propose to maximize the L21-norm based robust PCA objective, which is theoretically connected to the minimization of reconstruction error. More importantly, we propose the efficient non-greedy optimization algorithms to solve our objective and the more general L21-norm maximization problem with theoretically guaranteed convergence. Experimental results on real world data sets show the effectiveness of the proposed method for principal component analysis.

cs.LG↗

arXiv2015

Improved Spectral Clustering via Embedded Label Propagation

Spectral clustering is a key research topic in the field of machine learning and data mining. Most of the existing spectral clustering algorithms are built upon Gaussian Laplacian matrices, which are sensitive to parameters. We propose a novel parameter free, distance consistent Locally Linear Embedding. The proposed distance consistent LLE promises that edges between closer data points have greater weight.Furthermore, we propose a novel improved spectral clustering via embedded label propagation. Our algorithm is built upon two advancements of the state of the art:1) label propagation,which propagates a nodeś labels to neighboring nodes according to their proximity; and 2) manifold learning, which has been widely used in its capacity to leverage the manifold structure of data points. First we perform standard spectral clustering on original data and assign each cluster to k nearest data points. Next, we propagate labels through dense, unlabeled data regions. Extensive experiments with various datasets validate the superiority of the proposed algorithm compared to current state of the art spectral algorithms.

cs.LG↗

arXiv2014

A Convex Sparse PCA for Feature Analysis

Principal component analysis (PCA) has been widely applied to dimensionality reduction and data pre-processing for different applications in engineering, biology and social science. Classical PCA and its variants seek for linear projections of the original variables to obtain a low dimensional feature representation with maximal variance. One limitation is that it is very difficult to interpret the results of PCA. In addition, the classical PCA is vulnerable to certain noisy data. In this paper, we propose a convex sparse principal component analysis (CSPCA) algorithm and apply it to feature analysis. First we show that PCA can be formulated as a low-rank regression optimization problem. Based on the discussion, the l 2 , 1 -norm minimization is incorporated into the objective function to make the regression coefficients sparse, thereby robust to the outliers. In addition, based on the sparse model used in CSPCA, an optimal weight is assigned to each of the original feature, which in turn provides the output with good interpretability. With the output of our CSPCA, we can effectively analyze the importance of each feature under the PCA criteria. The objective function is convex, and we propose an iterative algorithm to optimize it. We apply the CSPCA algorithm to feature selection and conduct extensive experiments on six different benchmark datasets. Experimental results demonstrate that the proposed algorithm outperforms state-of-the-art unsupervised feature selection algorithms.

cs.LG↗

arXiv2012

An Iterative Locally Linear Embedding Algorithm

Local Linear embedding (LLE) is a popular dimension reduction method. In this paper, we first show LLE with nonnegative constraint is equivalent to the widely used Laplacian embedding. We further propose to iterate the two steps in LLE repeatedly to improve the results. Thirdly, we relax the kNN constraint of LLE and present a sparse similarity learning algorithm. The final Iterative LLE combines these three improvements. Extensive experiment results show that iterative LLE algorithm significantly improve both classification and clustering results.

cs.LG↗

arXiv2011

Continuous monitoring of membrane protein micro-domain association during cell signaling

Central to understanding membrane bound cell signaling is to quantify how the membrane ultra-structure consisting of transient spatial domains modulates signaling and how the signaling influences this ultra-structure. Yet, measuring the association of membrane proteins with domains in living, intact cells poses considerable challenges. Here, we describe a non-destructive method to quantify protein-lipid domain and protein cytoskeleton interactions in single, intact cells enabling continuous monitoring of the protein domains interaction over time during signaling.

physics.bio-ph↗

arXiv2009

Are Tensor Decomposition Solutions Unique? On the global convergence of HOSVD and ParaFac algorithms

For tensor decompositions such as HOSVD and ParaFac, the objective functions are nonconvex. This implies, theoretically, there exists a large number of local optimas: starting from different starting point, the iteratively improved solution will converge to different local solutions. This non-uniqueness present a stability and reliability problem for image compression and retrieval. In this paper, we present the results of a comprehensive investigation of this problem. We found that although all tensor decomposition algorithms fail to reach a unique global solution on random data and severely scrambled data; surprisingly however, on all real life several data sets (even with substantial scramble and occlusions), HOSVD always produce the unique global solution in the parameter region suitable to practical applications, while ParaFac produce non-unique solutions. We provide an eigenvalue based rule for the assessing the solution uniqueness.

cs.CV↗