arXiv subjects

Hui Jiang

Publications and source records attributed to Hui Jiang.

At least 91 records · Page 5Linked to original sources

arXiv2017

Characterization of Lifshitz transitions in topological nodal line semimetals

We introduce a two-band model of three-dimensional nodal line semimetals, the Fermi surface of which at half-filling may form various one-dimensional configurations of different topology. We study the symmetries and "drumhead" surface states of the model, and find that the transitions between different configurations, namely, the Lifshitz transitions, can be identified solely by the number of gap-closing points on some high-symmetry planes in the Brillouin zone. A global phase diagram of this model is also obtained accordingly. We then investigate the effect of some extra terms analogous to a two-dimensional Rashba-type spin-orbit coupling. The introduced extra terms open a gap for the nodal line semimetals and can be useful in engineering different topological insulating phases. We demonstrate that the behavior of surface Dirac cones in the resulting insulating system has a clear correspondence with the different configurations of the original nodal lines in the absence of the gap terms.

cond-mat.mes-hall↗

arXiv2017

Recurrent Neural Network-Based Sentence Encoder with Gated Attention for Natural Language Inference

The RepEval 2017 Shared Task aims to evaluate natural language understanding models for sentence representation, in which a sentence is represented as a fixed-length vector with neural networks and the quality of the representation is tested with a natural language inference task. This paper describes our system (alpha) that is ranked among the top in the Shared Task, on both the in-domain test set (obtaining a 74.9% accuracy) and on the cross-domain test set (also attaining a 74.9% accuracy), demonstrating that the model generalizes well to the cross-domain data. Our model is equipped with intra-sentence gated-attention composition which helps achieve a better performance. In addition to submitting our model to the Shared Task, we have also tested it on the Stanford Natural Language Inference (SNLI) dataset. We obtain an accuracy of 85.5%, which is the best reported result on SNLI when cross-sentence attention is not allowed, the same condition enforced in RepEval 2017.

cs.CL↗

arXiv2017

Minimizing Sum of Truncated Convex Functions and Its Applications

In this paper, we study a class of problems where the sum of truncated convex functions is minimized. In statistical applications, they are commonly encountered when $\ell_0$-penalized models are fitted and usually lead to NP-Hard non-convex optimization problems. In this paper, we propose a general algorithm for the global minimizer in low-dimensional settings. We also extend the algorithm to high-dimensional settings, where an approximate solution can be found efficiently. We introduce several applications where the sum of truncated convex functions is used, compare our proposed algorithm with other existing algorithms in simulation studies, and show its utility in edge-preserving image restoration on real data.

stat.CO↗

arXiv2017

Supervised Adversarial Networks for Image Saliency Detection

In the past few years, Generative Adversarial Network (GAN) became a prevalent research topic. By defining two convolutional neural networks (G-Network and D-Network) and introducing an adversarial procedure between them during the training process, GAN has ability to generate good quality images that look like natural images from a random vector. Besides image generation, GAN may have potential to deal with wide range of real world problems. In this paper, we follow the basic idea of GAN and propose a novel model for image saliency detection, which is called Supervised Adversarial Networks (SAN). Specifically, SAN also trains two models simultaneously: the G-Network takes natural images as inputs and generates corresponding saliency maps (synthetic saliency maps), and the D-Network is trained to determine whether one sample is a synthetic saliency map or ground-truth saliency map. However, different from GAN, the proposed method uses fully supervised learning to learn both G-Network and D-Network by applying class labels of the training set. Moreover, a novel kind of layer call conv-comparison layer is introduced into the D-Network to further improve the saliency performance by forcing the high-level feature of synthetic saliency maps and ground-truthes as similar as possible. Experimental results on Pascal VOC 2012 database show that the SAN model can generate high quality saliency maps for many complicate natural images.

cs.CV↗

arXiv2017

Enhanced LSTM for Natural Language Inference

Reasoning and inference are central to human and artificial intelligence. Modeling inference in human language is very challenging. With the availability of large annotated data (Bowman et al., 2015), it has recently become feasible to train neural network based inference models, which have shown to be very effective. In this paper, we present a new state-of-the-art result, achieving the accuracy of 88.6% on the Stanford Natural Language Inference Dataset. Unlike the previous top models that use very complicated network architectures, we first demonstrate that carefully designing sequential inference models based on chain LSTMs can outperform all previous models. Based on this, we further show that by explicitly considering recursive architectures in both local inference modeling and inference composition, we achieve additional improvement. Particularly, incorporating syntactic parsing information contributes to our best result---it further improves the performance even when added to the already very strong model.

cs.CL↗

arXiv2017

Exploring Question Understanding and Adaptation in Neural-Network-Based Question Answering

The last several years have seen intensive interest in exploring neural-network-based models for machine comprehension (MC) and question answering (QA). In this paper, we approach the problems by closely modelling questions in a neural network framework. We first introduce syntactic information to help encode questions. We then view and model different types of questions and the information shared among them as an adaptation task and proposed adaptation models for them. On the Stanford Question Answering Dataset (SQuAD), we show that these approaches can help attain better results over a competitive baseline.

cs.CL↗

arXiv2017

Fast Approximation of Small p-values in Permutation Tests by Partitioning the Permutations

Researchers in genetics and other life sciences commonly use permutation tests to evaluate differences between groups. Permutation tests have desirable properties, including exactness if data are exchangeable, and are applicable even when the distribution of the test statistic is analytically intractable. However, permutation tests can be computationally intensive. We propose both an asymptotic approximation and a resampling algorithm for quickly estimating small permutation p-values (e.g. $<10^{-6}$) for the difference and ratio of means in two-sample tests. Our methods are based on the distribution of test statistics within and across partitions of the permutations, which we define. In this article, we present our methods and demonstrate their use through simulations and an application to cancer genomic data. Through simulations, we find that our resampling algorithm is more computationally efficient than another leading alternative, particularly for extremely small p-values (e.g. $<10^{-30}$). Through application to cancer genomic data, we find that our methods can successfully identify up- and down-regulated genes. While we focus on the difference and ratio of means, we speculate that our approaches may work in other settings.

stat.CO↗

arXiv2017

Moderate deviations for bipower variation of general function and Hayashi-Yoshida estimators

We consider the moderate deviations behaviors for two (co-) volatility estima-tors: generalised bipower variation, Hayashi-Yoshida estimator. The results are obtained by using a new result about the moderate deviations principle for m-dependent random variables based on the Chen-Ledoux type condition. In the last decade there has been a considerable development of the asymptotic theory for processes observed at a high frequency. This was mainly motivated by financial applications , where the data, such as stock prices or currencies, are observed very frequently. As under the no-arbitrage assumptions price processes must follow a semimartingale, there was a need for probabilistic tools for functionals of semimartingales based on high frequency observations. Inspired by potential applications, probabilists started to develop limit theorems for semimartingales. Statisticians applied the asymptotic theory to analyze the path properties of discretely observed semimartingales: for the estimation of certain volatility functionals and realised jumps, or for performing various test procedures. We consider X t = (X 1,t , X 2,t) t$\in$[0,T ] a 2-dimensional semimartingale, defined on the filtred probability space ($Ω$, F , (F t) [0,T ] , P), of the form

math.PR↗

arXiv2016

Commonsense Knowledge Enhanced Embeddings for Solving Pronoun Disambiguation Problems in Winograd Schema Challenge

In this paper, we propose commonsense knowledge enhanced embeddings (KEE) for solving the Pronoun Disambiguation Problems (PDP). The PDP task we investigate in this paper is a complex coreference resolution task which requires the utilization of commonsense knowledge. This task is a standard first round test set in the 2016 Winograd Schema Challenge. In this task, traditional linguistic features that are useful for coreference resolution, e.g. context and gender information, are no longer effective anymore. Therefore, the KEE models are proposed to provide a general framework to make use of commonsense knowledge for solving the PDP problems. Since the PDP task doesn't have training data, the KEE models would be used during the unsupervised feature extraction process. To evaluate the effectiveness of the KEE models, we propose to incorporate various commonsense knowledge bases, including ConceptNet, WordNet, and CauseCom, into the KEE training process. We achieved the best performance by applying the proposed methods to the 2016 Winograd Schema Challenge. In addition, experiments conducted on the standard PDP task indicate that, the proposed KEE models could solve the PDP problems by achieving 66.7% accuracy, which is a new state-of-the-art performance.

cs.AI↗

arXiv2016

Erdős-Gallai-type results for total monochromatic connection of graphs

A graph is said to be {\it total-colored} if all the edges and the vertices of the graph are colored. A total-coloring of a graph is a {\it total monochromatically-connecting coloring} ({\it TMC-coloring}, for short) if any two vertices of the graph are connected by a path whose edges and internal vertices have the same color. For a connected graph $G$, the {\it total monochromatic connection number}, denoted by $tmc(G)$, is defined as the maximum number of colors used in a TMC-coloring of $G$. In this paper, we study two kinds of Erdős-Gallai-type problems for $tmc(G)$ and completely solve them.

math.CO↗

arXiv2016

Generating images with recurrent adversarial networks

Gatys et al. (2015) showed that optimizing pixels to match features in a convolutional network with respect reference image features is a way to render images of high visual quality. We show that unrolling this gradient-based optimization yields a recurrent computation that creates images by incrementally adding onto a visual "canvas". We propose a recurrent generative model inspired by this view, and show that it can be trained using adversarial training to generate very good image samples. We also propose a way to quantitatively compare adversarial networks by having the generators and discriminators of these networks compete against each other.

cs.LG↗

arXiv2016

Neural Networks Models for Entity Discovery and Linking

This paper describes the USTC_NELSLIP systems submitted to the Trilingual Entity Detection and Linking (EDL) track in 2016 TAC Knowledge Base Population (KBP) contests. We have built two systems for entity discovery and mention detection (MD): one uses the conditional RNNLM and the other one uses the attention-based encoder-decoder framework. The entity linking (EL) system consists of two modules: a rule based candidate generation and a neural networks probability ranking model. Moreover, some simple string matching rules are used for NIL clustering. At the end, our best system has achieved an F1 score of 0.624 in the end-to-end typed mention ceaf plus metric.

cs.CL↗

arXiv2016

A FOFE-based Local Detection Approach for Named Entity Recognition and Mention Detection

In this paper, we study a novel approach for named entity recognition (NER) and mention detection in natural language processing. Instead of treating NER as a sequence labelling problem, we propose a new local detection approach, which rely on the recent fixed-size ordinally forgetting encoding (FOFE) method to fully encode each sentence fragment and its left/right contexts into a fixed-size representation. Afterwards, a simple feedforward neural network is used to reject or predict entity label for each individual fragment. The proposed method has been evaluated in several popular NER and mention detection tasks, including the CoNLL 2003 NER task and TAC-KBP2015 and TAC-KBP2016 Tri-lingual Entity Discovery and Linking (EDL) tasks. Our methods have yielded pretty strong performance in all of these examined tasks. This local detection approach has shown many advantages over the traditional sequence labelling methods.

cs.CL↗

arXiv2016

Distraction-Based Neural Networks for Document Summarization

Distributed representation learned with neural networks has recently shown to be effective in modeling natural languages at fine granularities such as words, phrases, and even sentences. Whether and how such an approach can be extended to help model larger spans of text, e.g., documents, is intriguing, and further investigation would still be desirable. This paper aims to enhance neural network models for such a purpose. A typical problem of document-level modeling is automatic summarization, which aims to model documents in order to generate summaries. In this paper, we propose neural models to train computers not just to pay attention to specific regions and content of input documents with attention models, but also distract them to traverse between different content of a document so as to better grasp the overall meaning for summarization. Without engineering any features, we train the models on two large datasets. The models achieve the state-of-the-art performance, and they significantly benefit from the distraction modeling, particularly when input documents are long.

cs.CL↗

arXiv2016

A Unified Model for Differential Expression Analysis of RNA-seq Data via L1-Penalized Linear Regression

The RNA-sequencing (RNA-seq) is becoming increasingly popular for quantifying gene expression levels. Since the RNA-seq measurements are relative in nature, between-sample normalization of counts is an essential step in differential expression (DE) analysis. The normalization of existing DE detection algorithms is ad hoc and performed once for all prior to DE detection, which may be suboptimal since ideally normalization should be based on non-DE genes only and thus coupled with DE detection. We propose a unified statistical model for joint normalization and DE detection of log-transformed RNA-seq data. Sample-specific normalization factors are modeled as unknown parameters in the gene-wise linear models and jointly estimated with the regression coefficients. By imposing sparsity-inducing L1 penalty (or mixed L1/L2-norm for multiple treatment conditions) on the regression coefficients, we formulate the problem as a penalized least-squares regression problem and apply the augmented lagrangian method to solve it. Simulation studies show that the proposed model and algorithms outperform existing methods in terms of detection power and false-positive rate when more than half of the genes are differentially expressed and/or when the up- and down-regulated genes among DE genes are unbalanced in amount.

stat.ME↗

arXiv2016

Learning Convolutional Neural Networks using Hybrid Orthogonal Projection and Estimation

Convolutional neural networks (CNNs) have yielded the excellent performance in a variety of computer vision tasks, where CNNs typically adopt a similar structure consisting of convolution layers, pooling layers and fully connected layers. In this paper, we propose to apply a novel method, namely Hybrid Orthogonal Projection and Estimation (HOPE), to CNNs in order to introduce orthogonality into the CNN structure. The HOPE model can be viewed as a hybrid model to combine feature extraction using orthogonal linear projection with mixture models. It is an effective model to extract useful information from the original high-dimension feature vectors and meanwhile filter out irrelevant noises. In this work, we present three different ways to apply the HOPE models to CNNs, i.e., {\em HOPE-Input}, {\em single-HOPE-Block} and {\em multi-HOPE-Blocks}. For {\em HOPE-Input} CNNs, a HOPE layer is directly used right after the input to de-correlate high-dimension input feature vectors. Alternatively, in {\em single-HOPE-Block} and {\em multi-HOPE-Blocks} CNNs, we consider to use HOPE layers to replace one or more blocks in the CNNs, where one block may include several convolutional layers and one pooling layer. The experimental results on both Cifar-10 and Cifar-100 data sets have shown that the orthogonal constraints imposed by the HOPE layers can significantly improve the performance of CNNs in these image classification tasks (we have achieved one of the best performance when image augmentation has not been applied, and top 5 performance with image augmentation).

cs.NE↗

arXiv2016

Unit-free and robust detection of differential expression from RNA-Seq data

Ultra high-throughput sequencing of transcriptomes (RNA-Seq) is a widely used method for quantifying gene expression levels due to its low cost, high accuracy and wide dynamic range for detection. However, the nature of RNA-Seq makes it nearly impossible to provide absolute measurements of transcript abundances. Several units or data summarization methods for transcript quantification have been proposed in the past to account for differences in transcript lengths and sequencing depths across different genes and different samples. Nevertheless, further between-sample normalization is still needed for reliable detection of differentially expressed genes. In this paper we propose a unified statistical model for joint detection of differential gene expression and between-sample normalization. Our method is independent of the unit in which gene expression levels are summarized. We also introduce an efficient algorithm for model fitting. Due to the L0-penalized likelihood used in our model, it is able to reliably normalize the data and detect differential gene expression in some cases when more than $50\%$ of the genes are differentially expressed in an asymmetric manner. We compare our method with existing methods using simulated and real data sets.

stat.ME↗

arXiv2016

Probabilistic Reasoning via Deep Learning: Neural Association Models

In this paper, we propose a new deep learning approach, called neural association model (NAM), for probabilistic reasoning in artificial intelligence. We propose to use neural networks to model association between any two events in a domain. Neural networks take one event as input and compute a conditional probability of the other event to model how likely these two events are to be associated. The actual meaning of the conditional probabilities varies between applications and depends on how the models are trained. In this work, as two case studies, we have investigated two NAM structures, namely deep neural networks (DNN) and relation-modulated neural nets (RMNN), on several probabilistic reasoning tasks in AI, including recognizing textual entailment, triple classification in multi-relational knowledge bases and commonsense reasoning. Experimental results on several popular datasets derived from WordNet, FreeBase and ConceptNet have all demonstrated that both DNNs and RMNNs perform equally well and they can significantly outperform the conventional methods available for these reasoning tasks. Moreover, compared with DNNs, RMNNs are superior in knowledge transfer, where a pre-trained model can be quickly extended to an unseen relation after observing only a few training samples. To further prove the effectiveness of the proposed models, in this work, we have applied NAMs to solving challenging Winograd Schema (WS) problems. Experiments conducted on a set of WS problems prove that the proposed models have the potential for commonsense reasoning.

cs.AI↗