• Publications
  • Influence
Residual Attention Network for Image Classification
TLDR
The proposed Residual Attention Network is a convolutional neural network using attention mechanism which can incorporate with state-of-art feed forward network architecture in an end-to-end training fashion and can be easily scaled up to hundreds of layers. Expand
Label Propagation through Linear Neighborhoods
TLDR
A novel graph-based semi supervised learning approach is proposed based on a linear neighborhood model, which assumes that each data point can be linearly reconstructed from its neighborhood, and can propagate the labels from the labeled points to the whole data set using these linear neighborhoods with sufficient smoothness. Expand
Late Mesozoic volcanism in the Great Xing'an Range (NE China): Timing and implications for the dynamic setting of NE Asia
Abstract Mesozoic volcanism is widespread throughout northeastern China, but precise geochronological constraints were previously lacking. Twenty samples, including basalt and basaltic andesites,Expand
Label Propagation through Linear Neighborhoods
TLDR
A novel graph-based semi supervised learning approach is proposed based on a linear neighborhood model, which assumes that each data point can be linearly reconstructed from its neighborhood, and can propagate the labels from the labeled points to the whole data set using these linear neighborhoods with sufficient smoothness. Expand
Generating a Prion with Bacterially Expressed Recombinant Prion Protein
TLDR
Recombinant prion protein recapitulates the characteristics of the infectious agent in prion disease and is resistant to proteinase-K, but also shows infectivity in cultured cells and causes rapid disease progression in wild-type mice, yielding both the behavioral and the neuropathological symptoms. Expand
Differentially Private Generative Adversarial Network
TLDR
This paper proposes a differentially private GAN (DPGAN) model, in which it is demonstrated that the method can generate high quality data points at a reasonable privacy level by adding carefully designed noise to gradients during the learning procedure. Expand
Community discovery using nonnegative matrix factorization
TLDR
This paper investigates another important issue, community discovery, in network analysis, and chooses Nonnegative Matrix Factorization (NMF) as a tool to find the communities because of its powerful interpretability and close relationship between clustering methods. Expand
Patient Subtyping via Time-Aware LSTM Networks
TLDR
A patient subtyping model is proposed that leverages the proposed T-LSTM in an auto-encoder to learn a powerful single representation for sequential records of patients, which is then used to cluster patients into clinical subtypes. Expand
Composite hashing with multiple information sources
TLDR
The focus of the new research problem is to design an algorithm for incorporating the features from different information sources into the binary hashing codes efficiently and effectively, and to propose an algorithm CHMIS-AW (CHMIS with Adjusted Weights) for learning the codes. Expand
A chaperone cascade sorts proteins for posttranslational membrane insertion into the endoplasmic reticulum.
TLDR
The composition of a conserved multiprotein TMD recognition complex (TRC) is revealed and it is shown that distinct TRC subunits recognize the two types of TMD signals. Expand
...
1
2
3
4
5
...