arXiv subjects

Yang Feng

Publications and source records attributed to Yang Feng.

At least 253 records · Page 14Linked to original sources

arXiv2018

Speeding Up Neural Machine Translation Decoding by Cube Pruning

Although neural machine translation has achieved promising results, it suffers from slow translation speed. The direct consequence is that a trade-off has to be made between translation quality and speed, thus its performance can not come into full play. We apply cube pruning, a popular technique to speed up dynamic programming, into neural machine translation to speed up the translation. To construct the equivalence class, similar target hidden states are combined, leading to less RNN expansion operations on the target side and less \$\mathrm{softmax}\$ operations over the large target vocabulary. The experiments show that, at the same or even better translation quality, our method can translate faster compared with naive beam search by \$3.3\times\$ on GPUs and \$3.5\times\$ on CPUs.

cs.CL↗

arXiv2018

Video Re-localization

Many methods have been developed to help people find the video contents they want efficiently. However, there are still some unsolved problems in this area. For example, given a query video and a reference video, how to accurately localize a segment in the reference video such that the segment semantically corresponds to the query video? We define a distinctively new task, namely \textbf{video re-localization}, to address this scenario. Video re-localization is an important emerging technology implicating many applications, such as fast seeking in videos, video copy detection, video surveillance, etc. Meanwhile, it is also a challenging research task because the visual appearance of a semantic concept in videos can have large variations. The first hurdle to clear for the video re-localization task is the lack of existing datasets. It is labor expensive to collect pairs of videos with semantic coherence or correspondence and label the corresponding segments. We first exploit and reorganize the videos in ActivityNet to form a new dataset for video re-localization research, which consists of about 10,000 videos of diverse visual appearances associated with localized boundary information. Subsequently, we propose an innovative cross gated bilinear matching model such that every time-step in the reference video is matched against the attentively weighted query video. Consequently, the prediction of the starting and ending time is formulated as a classification problem based on the matching results. Extensive experimental results show that the proposed method outperforms the competing methods. Our code is available at: https://github.com/fengyang0317/video_reloc.

cs.CV↗

arXiv2018

Quantum anomalous Hall multilayers grown by molecular beam epitaxy

Quantum anomalous Hall (QAH) effect is a quantum Hall effect that occurs without the need of external magnetic field. A system composed of multiple parallel QAH layers is an effective high Chern number QAH insulator and the key to the applications of the dissipationless chiral edge channels in low energy consumption electronics. Such a QAH multilayer can also be engineered into other exotic topological phases such as a magnetic Weyl semimetal with only one pair of Weyl points. This work reports the first experimental realization of QAH multilayers in the superlattices composed of magnetically doped (Bi,Sb)$_2$Te$_3$ topological insulator and CdSe normal insulator layers grown by molecular beam epitaxy. The obtained multilayer samples show quantized Hall resistance $h/Ne$$^2$, where $h$ is the Planck's constant, $e$ is the elementary charge and $N$ is the number of the magnetic topological insulator layers, resembling a high Chern number QAH insulator.

cond-mat.mtrl-sci↗

arXiv2018

Information-Propogation-Enhanced Neural Machine Translation by Relation Model

Even though sequence-to-sequence neural machine translation (NMT) model have achieved state-of-art performance in the recent fewer years, but it is widely concerned that the recurrent neural network (RNN) units are very hard to capture the long-distance state information, which means RNN can hardly find the feature with long term dependency as the sequence becomes longer. Similarly, convolutional neural network (CNN) is introduced into NMT for speeding recently, however, CNN focus on capturing the local feature of the sequence; To relieve this issue, we incorporate a relation network into the standard encoder-decoder framework to enhance information-propogation in neural network, ensuring that the information of the source sentence can flow into the decoder adequately. Experiments show that proposed framework outperforms the statistical MT model and the state-of-art NMT model significantly on two data sets with different scales.

cs.CL↗

arXiv2018

Refining Source Representations with Relation Networks for Neural Machine Translation

Although neural machine translation (NMT) with the encoder-decoder framework has achieved great success in recent times, it still suffers from some drawbacks: RNNs tend to forget old information which is often useful and the encoder only operates through words without considering word relationship. To solve these problems, we introduce a relation networks (RN) into NMT to refine the encoding representations of the source. In our method, the RN first augments the representation of each source word with its neighbors and reasons all the possible pairwise relations between them. Then the source representations and all the relations are fed to the attention module and the decoder together, keeping the main encoder-decoder architecture unchanged. Experiments on two Chinese-to-English data sets in different scales both show that our method can outperform the competitive baselines significantly.

cs.CL↗

arXiv2018

Partial Distance Correlation Screening for High Dimensional Time Series

High dimensional time series datasets are becoming increasingly common in various fields such as economics, finance, meteorology, and neuroscience. Given this ubiquity of time series data, it is surprising that very few works on variable screening discuss the time series setting, and even fewer works have developed methods which utilize the unique features of time series data. This paper introduces several model free screening methods based on the partial distance correlation and developed specifically to deal with time dependent data. Methods are developed both for univariate models, such as nonlinear autoregressive models with exogenous predictors (NARX), and multivariate models such as linear or nonlinear VAR models. Sure screening properties are proved for our methods, which depend on the moment conditions, and the strength of dependence in the response and covariate processes, amongst other factors. Dependence is quantified by functional dependence measures (Wu [Proc. Natl. Acad. Sci. USA 102 (2005) 14150-14154]) and $β$-mixing coefficients, and the results rely on the use of Nagaev and Rosenthal type inequalities for dependent random variables. Finite sample performance of our methods is shown through extensive simulation studies, and we include an application to macroeconomic forecasting.

stat.ME↗

arXiv2018

Large-Scale Model Selection with Misspecification

Model selection is crucial to high-dimensional learning and inference for contemporary big data applications in pinpointing the best set of covariates among a sequence of candidate interpretable models. Most existing work assumes implicitly that the models are correctly specified or have fixed dimensionality. Yet both features of model misspecification and high dimensionality are prevalent in practice. In this paper, we exploit the framework of model selection principles in misspecified models originated in Lv and Liu (2014) and investigate the asymptotic expansion of Bayesian principle of model selection in the setting of high-dimensional misspecified models. With a natural choice of prior probabilities that encourages interpretability and incorporates Kullback-Leibler divergence, we suggest the high-dimensional generalized Bayesian information criterion with prior probability (HGBIC_p) for large-scale model selection with misspecification. Our new information criterion characterizes the impacts of both model misspecification and high dimensionality on model selection. We further establish the consistency of covariance contrast matrix estimation and the model selection consistency of HGBIC_p in ultra-high dimensions under some mild regularity conditions. The advantages of our new method are supported by numerical studies.

stat.ME↗

arXiv2018

Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications

To ensure undisrupted business, large Internet companies need to closely monitor various KPIs (e.g., Page Views, number of online users, and number of orders) of its Web applications, to accurately detect anomalies and trigger timely troubleshooting/mitigation. However, anomaly detection for these seasonal KPIs with various patterns and data quality has been a great challenge, especially without labels. In this paper, we proposed Donut, an unsupervised anomaly detection algorithm based on VAE. Thanks to a few of our key techniques, Donut greatly outperforms a state-of-arts supervised ensemble approach and a baseline VAE approach, and its best F-scores range from 0.75 to 0.9 for the studied KPIs from a top global Internet company. We come up with a novel KDE interpretation of reconstruction for Donut, making it the first VAE-based anomaly detection algorithm with solid theoretical explanation.

cs.LG↗

arXiv2018

The restricted consistency property of leave-$n_v$-out cross-validation for high-dimensional variable selection

Cross-validation (CV) methods are popular for selecting the tuning parameter in the high-dimensional variable selection problem. We show the mis-alignment of the CV is one possible reason of its over-selection behavior. To fix this issue, we propose a version of leave-$n_v$-out cross-validation (CV($n_v$)), for selecting the optimal model among the restricted candidate model set for high-dimensional generalized linear models. By using the same candidate model sequence and a proper order of construction sample size $n_c$ in each CV split, CV($n_v$) avoids the potential hurdles in developing theoretical properties. CV($n_v$) is shown to enjoy the restricted model selection consistency property under mild conditions. Extensive simulations and real data analysis support the theoretical results and demonstrate the performances of CV($n_v$) in terms of both model selection and prediction.

stat.ME↗

arXiv2017

Nonparametric Independence Screening via Favored Smoothing Bandwidth

We propose a flexible nonparametric regression method for ultrahigh-dimensional data. As a first step, we propose a fast screening method based on the favored smoothing bandwidth of the marginal local constant regression. Then, an iterative procedure is developed to recover both the important covariates and the regression function. Theoretically, we prove that the favored smoothing bandwidth based screening possesses the model selection consistency property. Simulation studies as well as real data analysis show the competitive performance of the new procedure.

stat.ME↗

arXiv2017

How Polarized Have We Become? A Multimodal Classification of Trump Followers and Clinton Followers

Polarization in American politics has been extensively documented and analyzed for decades, and the phenomenon became all the more apparent during the 2016 presidential election, where Trump and Clinton depicted two radically different pictures of America. Inspired by this gaping polarization and the extensive utilization of Twitter during the 2016 presidential campaign, in this paper we take the first step in measuring polarization in social media and we attempt to predict individuals' Twitter following behavior through analyzing ones' everyday tweets, profile images and posted pictures. As such, we treat polarization as a classification problem and study to what extent Trump followers and Clinton followers on Twitter can be distinguished, which in turn serves as a metric of polarization in general. We apply LSTM to processing tweet features and we extract visual features using the VGG neural network. Integrating these two sets of features boosts the overall performance. We are able to achieve an accuracy of 69%, suggesting that the high degree of polarization recorded in the literature has started to manifest itself in social media as well.

cs.SI↗

arXiv2017

Neyman-Pearson (NP) classification algorithms and NP receiver operating characteristics (NP-ROC)

In many binary classification applications such as disease diagnosis and spam detection, practitioners often face great needs to control type I errors (i.e., the conditional probability of misclassifying a class 0 observation as class 1) so that it remains below a desired threshold. To address this need, the Neyman-Pearson (NP) classification paradigm is a natural choice; it minimizes type II error (i.e., the conditional probability of misclassifying a class 1 observation as class 0) while enforcing an upper bound, $α$, on the type I error. Although the NP paradigm has a century-long history in hypothesis testing, it has not been well recognized and implemented in classification schemes. Common practices that directly limit the empirical type I error to no more than $α$ do not satisfy the type I error control objective because the resulting classifiers are still likely to have type I errors much larger than $α$. As a result, the NP paradigm has not been properly implemented for many classification scenarios in practice. In this work, we develop the first umbrella algorithm that implements the NP paradigm for all scoring-type classification methods, including popular methods such as logistic regression, support vector machines and random forests. Powered by this umbrella algorithm, we propose a novel graphical tool for NP classification methods: NP receiver operating characteristic (NP-ROC) bands, motivated by the popular receiver operating characteristic (ROC) curves. NP-ROC bands will help choose $α$ in a data adaptive way and compare different NP classifiers. We demonstrate the use and properties of the NP umbrella algorithm and NP-ROC bands, available in the R package nproc, through simulation and real data case studies.

stat.ME↗

arXiv2017

Regularization after retention in ultrahigh dimensional linear regression models

In ultrahigh dimensional setting, independence screening has been both theoretically and empirically proved a useful variable selection framework with low computation cost. In this work, we propose a two-step framework by using marginal information in a different perspective from independence screening. In particular, we retain significant variables rather than screening out irrelevant ones. The new method is shown to be model selection consistent in the ultrahigh dimensional linear regression model. To improve the finite sample performance, we then introduce a three-step version and characterize its asymptotic behavior. Simulations and real data analysis show advantages of our method over independence screening and its iterative variants in certain regimes.

stat.ME↗

arXiv2017

Memory-augmented Neural Machine Translation

Neural machine translation (NMT) has achieved notable success in recent times, however it is also widely recognized that this approach has limitations with handling infrequent words and word pairs. This paper presents a novel memory-augmented NMT (M-NMT) architecture, which stores knowledge about how words (usually infrequently encountered ones) should be translated in a memory and then utilizes them to assist the neural model. We use this memory mechanism to combine the knowledge learned from a conventional statistical machine translation system and the rules learned by an NMT system, and also propose a solution for out-of-vocabulary (OOV) words based on this framework. Our experiments on two Chinese-English translation tasks demonstrated that the M-NMT architecture outperformed the NMT baseline by $9.0$ and $2.7$ BLEU points on the two tasks, respectively. Additionally, we found this architecture resulted in a much more effective OOV treatment compared to competitive methods.

cs.CL↗

arXiv2017

Magnetic quantum phase transition in Cr-doped Bi2(SexTe1-x)3 driven by the Stark effect

The interplay between magnetism and topology, as exemplified in the magnetic skyrmion systems, has emerged as a rich playground for finding novel quantum phenomena and applications in future information technology. Magnetic topological insulators (TI) have attracted much recent attention, especially after the experimental realization of quantum anomalous Hall effect. Future applications of magnetic TI hinge on the accurate manipulation of magnetism and topology by external perturbations, preferably with a gate electric field. In this work, we investigate the magneto transport properties of Cr doped Bi2(SexTe1-x)3 TI across the topological quantum critical point (QCP). We find that the external gate voltage has negligible effect on the magnetic order for samples far away from the topological QCP. But for the sample near the QCP, we observe a ferromagnetic (FM) to paramagnetic (PM) phase transition driven by the gate electric field. Theoretical calculations show that a perpendicular electric field causes a shift of electronic energy levels due to the Stark effect, which induces a topological quantum phase transition and consequently a magnetic phase transition. The in situ electrical control of the topological and magnetic properties of TI shed important new lights on future topological electronic or spintronic device applications.

cond-mat.mtrl-sci↗

arXiv2017

Collaborative Learning for Language and Speaker Recognition

This paper presents a unified model to perform language and speaker recognition simultaneously and altogether. The model is based on a multi-task recurrent neural network where the output of one task is fed as the input of the other, leading to a collaborative learning framework that can improve both language and speaker recognition by borrowing information from each other. Our experiments demonstrated that the multi-task model outperforms the task-specific models on both tasks.

cs.SD↗

arXiv2017

Flexible and Creative Chinese Poetry Generation Using Neural Memory

It has been shown that Chinese poems can be successfully generated by sequence-to-sequence neural models, particularly with the attention mechanism. A potential problem of this approach, however, is that neural models can only learn abstract rules, while poem generation is a highly creative process that involves not only rules but also innovations for which pure statistical models are not appropriate in principle. This work proposes a memory-augmented neural model for Chinese poem generation, where the neural model and the augmented memory work together to balance the requirements of linguistic accordance and aesthetic innovation, leading to innovative generations that are still rule-compliant. In addition, it is found that the memory mechanism provides interesting flexibility that can be used to generate poems with different styles.

cs.AI↗

arXiv2017

Memory Visualization for Gated Recurrent Neural Networks in Speech Recognition

Recurrent neural networks (RNNs) have shown clear superiority in sequence modeling, particularly the ones with gated units, such as long short-term memory (LSTM) and gated recurrent unit (GRU). However, the dynamic properties behind the remarkable performance remain unclear in many applications, e.g., automatic speech recognition (ASR). This paper employs visualization techniques to study the behavior of LSTM and GRU when performing speech recognition tasks. Our experiments show some interesting patterns in the gated memory, and some of them have inspired simple yet effective modifications on the network structure. We report two of such modifications: (1) lazy cell update in LSTM, and (2) shortcut connections for residual learning. Both modifications lead to more comprehensible and powerful networks.

cs.LG↗