arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 1,549 records · Page 86Linked to original sources

arXiv2019

Trimming Mobile Applications for Bandwidth-Challenged Networks in Developing Regions

Despite continuous efforts to build and update network infrastructure, mobile devices in developing regions continue to be constrained by limited bandwidth. Unfortunately, this coincides with a period of unprecedented growth in the size of mobile applications. Thus it is becoming prohibitively expensive for users in developing regions to download and update mobile apps critical to their economic and educational development. Unchecked, these trends can further contribute to a large and growing global digital divide. Our goal is to better understand the source of this rapid growth in mobile app code size, whether it is reflective of new functionality, and identify steps that can be taken to make existing mobile apps more friendly bandwidth constrained mobile networks. We hypothesize that much of this growth in mobile apps is due to poor resource/code management, and do not reflect proportional increases in functionality. Our hypothesis is partially validated by mini-programs, apps with extremely small footprints gaining popularity in Chinese mobile networks. Here, we use functionally equivalent pairs of mini-programs and Android apps to identify potential sources of "bloat," inefficient uses of code or resources that contribute to large package sizes. We analyze a large sample of popular Android apps and quantify instances of code and resource bloat. We develop techniques for automated code and resource trimming, and successfully validate them on a large set of Android apps. We hope our results will lead to continued efforts to streamline mobile apps, making them easier to access and maintain for users in developing regions.

cs.SE↗

arXiv2019

A novel generative reverse net assisted evolution algorithm for expensive-computational optimizations

Simulation-based optimization is a useful method for practical design problems. However, it is difficult for complicated problems due to expensive-computational costs. A popular way to overcome this issue is to use a surrogate model to save the cost. Nevertheless, limited design parameters those are input to traditional surrogate models can difficultly represent the whole design problem, which might result in unexpected errors. In this study, physical cloud images from simulations are employed and attempted to construct the surrogate model. Simultaneously, based on the strong pattern recognition and generation abilities of deep learning models, a novel Generative Reverse Net assisted Evolution Algorithm (GRN-EA) is proposed for expensive-design problems. In this study, a numerical example of a Variable-Stiffness (VS) composite hole-plate is employed to obtain the optimal distribution of the curved fiber. Moreover, to evaluate the performance of GRN-EA in practical engineering problems, a more complex sheet-forming case is optimized. According to the two problems, some expensive-simulations can be well handled in this study.

eess.SP↗

arXiv2019

"How do urban incidents affect traffic speed?" A Deep Graph Convolutional Network for Incident-driven Traffic Speed Prediction

Accurate traffic speed prediction is an important and challenging topic for transportation planning. Previous studies on traffic speed prediction predominately used spatio-temporal and context features for prediction. However, they have not made good use of the impact of urban traffic incidents. In this work, we aim to make use of the information of urban incidents to achieve a better prediction of traffic speed. Our incident-driven prediction framework consists of three processes. First, we propose a critical incident discovery method to discover urban traffic incidents with high impact on traffic speed. Second, we design a binary classifier, which uses deep learning methods to extract the latent incident impact features from the middle layer of the classifier. Combining above methods, we propose a Deep Incident-Aware Graph Convolutional Network (DIGC-Net) to effectively incorporate urban traffic incident, spatio-temporal, periodic and context features for traffic speed prediction. We conduct experiments on two real-world urban traffic datasets of San Francisco and New York City. The results demonstrate the superior performance of our model compare to the competing benchmarks.

cs.LG↗

arXiv2019

Speaker detection in the wild: Lessons learned from JSALT 2019

This paper presents the problems and solutions addressed at the JSALT workshop when using a single microphone for speaker detection in adverse scenarios. The main focus was to tackle a wide range of conditions that go from meetings to wild speech. We describe the research threads we explored and a set of modules that was successful for these scenarios. The ultimate goal was to explore speaker detection; but our first finding was that an effective diarization improves detection, and not having a diarization stage impoverishes the performance. All the different configurations of our research agree on this fact and follow a main backbone that includes diarization as a previous stage. With this backbone, we analyzed the following problems: voice activity detection, how to deal with noisy signals, domain mismatch, how to improve the clustering; and the overall impact of previous stages in the final speaker detection. In this paper, we show partial results for speaker diarizarion to have a better understanding of the problem and we present the final results for speaker detection.

eess.AS↗

arXiv2019

Domain-Aware Dynamic Networks

Deep neural networks with more parameters and FLOPs have higher capacity and generalize better to diverse domains. But to be deployed on edge devices, the model's complexity has to be constrained due to limited compute resource. In this work, we propose a method to improve the model capacity without increasing inference-time complexity. Our method is based on an assumption of data locality: for an edge device, within a short period of time, the input data to the device are sampled from a single domain with relatively low diversity. Therefore, it is possible to utilize a specialized, low-complexity model to achieve good performance in that input domain. To leverage this, we propose Domain-aware Dynamic Network (DDN), which is a high-capacity dynamic network in which each layer contains multiple weights. During inference, based on the input domain, DDN dynamically combines those weights into one single weight that specializes in the given domain. This way, DDN can keep the inference-time complexity low but still maintain a high capacity. Experiments show that without increasing the parameters, FLOPs, and actual latency, DDN achieves up to 2.6\% higher AP50 than a static network on the BDD100K object-detection benchmark.

cs.LG↗

arXiv2019

On the Properties of the Effective Jarlskog Invariant for Three-flavor Neutrino Oscillations in Matter

In this paper, we show that the ratio of the effective Jarlskog invariant $\widetilde{\cal J}$ for leptonic CP violation in three-flavor neutrino oscillations in matter to its counterpart ${\cal J}$ in vacuum $\widetilde{\cal J}/{\cal J} \approx 1/(\hat{C}^{}_{12} \hat{C}^{}_{13})$ holds as an excellent approximation, where $\hat{C}^{}_{12} \equiv \sqrt{1 - 2 \hat{A}^{}_* \cos 2θ^{}_{12} + \hat{A}^2_*}$ with $\hat{A}^{}_* \equiv a\cos^2 θ^{}_{13}/Δ^{}_{21}$ and $\hat{C}^{}_{13} \equiv \sqrt{1 - 2 A^{}_{\rm c} \cos 2θ^{}_{13} + A^2_{\rm c}}$ with $A^{}_{\rm c} \equiv a/Δ^{}_{\rm c}$. Here $Δ^{}_{ij} \equiv m^2_i - m^2_j$ (for $ij = 21, 31, 32$) stand for the neutrino mass-squared differences in vacuum and $θ^{}_{ij}$ (for $ij = 12, 13, 23$) are the neutrino mixing angles in vacuum, while $Δ^{}_{\rm c} \equiv Δ^{}_{31}\cos^2θ^{}_{12} + Δ^{}_{32} \sin^2 θ^{}_{12}$ and the matter parameter $a \equiv 2\sqrt{2}G^{}_{\rm F} N^{}_e E$ are defined. This result has been explicitly derived by improving the previous analytical solutions to the renormalization-group equations of effective neutrino masses and mixing parameters in matter. Furthermore, as a practical application, such a simple analytical formula has been implemented to understand the existence and location of the extrema of $\widetilde{\cal J}$.

hep-ph↗

arXiv2019

Transformation of low-quality device-recorded speech to high-quality speech using improved SEGAN model

Nowadays vast amounts of speech data are recorded from low-quality recorder devices such as smartphones, tablets, laptops, and medium-quality microphones. The objective of this research was to study the automatic generation of high-quality speech from such low-quality device-recorded speech, which could then be applied to many speech-generation tasks. In this paper, we first introduce our new device-recorded speech dataset then propose an improved end-to-end method for automatically transforming the low-quality device-recorded speech into professional high-quality speech. Our method is an extension of a generative adversarial network (GAN)-based speech enhancement model called speech enhancement GAN (SEGAN), and we present two modifications to make model training more robust and stable. Finally, from a large-scale listening test, we show that our method can significantly enhance the quality of device-recorded speech signals.

cs.SD↗

arXiv2019

Transferring neural speech waveform synthesizers to musical instrument sounds generation

Recent neural waveform synthesizers such as WaveNet, WaveGlow, and the neural-source-filter (NSF) model have shown good performance in speech synthesis despite their different methods of waveform generation. The similarity between speech and music audio synthesis techniques suggests interesting avenues to explore in terms of the best way to apply speech synthesizers in the music domain. This work compares three neural synthesizers used for musical instrument sounds generation under three scenarios: training from scratch on music data, zero-shot learning from the speech domain, and fine-tuning-based adaptation from the speech to the music domain. The results of a large-scale perceptual test demonstrated that the performance of three synthesizers improved when they were pre-trained on speech data and fine-tuned on music data, which indicates the usefulness of knowledge from speech data for music audio generation. Among the synthesizers, WaveGlow showed the best potential in zero-shot learning while NSF performed best in the other scenarios and could generate samples that were perceptually close to natural audio.

eess.AS↗

arXiv2019

Adaptive Activation Network and Functional Regularization for Efficient and Flexible Deep Multi-Task Learning

Multi-task learning (MTL) is a common paradigm that seeks to improve the generalization performance of task learning by training related tasks simultaneously. However, it is still a challenging problem to search the flexible and accurate architecture that can be shared among multiple tasks. In this paper, we propose a novel deep learning model called Task Adaptive Activation Network (TAAN) that can automatically learn the optimal network architecture for MTL. The main principle of TAAN is to derive flexible activation functions for different tasks from the data with other parameters of the network fully shared. We further propose two functional regularization methods that improve the MTL performance of TAAN. The improved performance of both TAAN and the regularization methods is demonstrated by comprehensive experiments.

cs.LG↗

arXiv2019

Neural source-filter waveform models for statistical parametric speech synthesis

Neural waveform models such as WaveNet have demonstrated better performance than conventional vocoders for statistical parametric speech synthesis. As an autoregressive (AR) model, WaveNet is limited by a slow sequential waveform generation process. Some new models that use the inverse-autoregressive flow (IAF) can generate a whole waveform in a one-shot manner. However, these IAF-based models require sequential transformation during training, which severely slows down the training speed. Other models such as Parallel WaveNet and ClariNet bring together the benefits of AR and IAF-based models and train an IAF model by transferring the knowledge from a pre-trained AR teacher to an IAF student without any sequential transformation. However, both models require additional training criteria, and their implementation is prohibitively complicated. We propose a framework for neural source-filter (NSF) waveform modeling without AR nor IAF-based approaches. This framework requires only three components for waveform generation: a source module that generates a sine-based signal as excitation, a non-AR dilated-convolution-based filter module that transforms the excitation into a waveform, and a conditional module that pre-processes the acoustic features for the source and filer modules. This framework minimizes spectral-amplitude distances for model training, which can be efficiently implemented by using short-time Fourier transform routines. Under this framework, we designed three NSF models and compared them with WaveNet. It was demonstrated that the NSF models generated waveforms at least 100 times faster than WaveNet, and the quality of the synthetic speech from the best NSF model was better than or equally good as that from WaveNet.

eess.AS↗

arXiv2019

Disparity-preserved Deep Cross-platform Association for Cross-platform Video Recommendation

Cross-platform recommendation aims to improve recommendation accuracy through associating information from different platforms. Existing cross-platform recommendation approaches assume all cross-platform information to be consistent with each other and can be aligned. However, there remain two unsolved challenges: i) there exist inconsistencies in cross-platform association due to platform-specific disparity, and ii) data from distinct platforms may have different semantic granularities. In this paper, we propose a cross-platform association model for cross-platform video recommendation, i.e., Disparity-preserved Deep Cross-platform Association (DCA), taking platform-specific disparity and granularity difference into consideration. The proposed DCA model employs a partially-connected multi-modal autoencoder, which is capable of explicitly capturing platform-specific information, as well as utilizing nonlinear mapping functions to handle granularity differences. We then present a cross-platform video recommendation approach based on the proposed DCA model. Extensive experiments for our cross-platform recommendation framework on real-world dataset demonstrate that the proposed DCA model significantly outperform existing cross-platform recommendation methods in terms of various evaluation metrics.

cs.IR↗

arXiv2019

Layer Dependence of Stacking Order in Nonencapsulated Few-layer CrI3

Long-range magnetic orders in atomically thin ferromagnetic CrI3 give rise to new fascinating physics and application perspectives. The physical properties of two-dimensional (2D) ferromagnetism CrI3 are significantly influenced by interlayer spacing and stacking order, which are sensitive to the hydrostatic pressure and external environments. However, there remains debate on the stacking order at low temperature. Here, we study the interlayer coupling and stacking order of non-encapsulated 2-5 layer and bulk CrI3 at 10 K by Raman spectroscopy; demonstrate a rhombohedral stacking in both antiferromagnetic and ferromagnetic CrI3. The opposite helicity dependence of Ag and Eg modes arising from phonon symmetry further validate the rhombohedral stacking. An anomalous temperature-dependent behavior is observed due to spin-phonon coupling below 60 K. Our work provides insights into the interlayer coupling and stacking orders of 2D ferromagnetic materials.

cond-mat.mtrl-sci↗

arXiv2019

Initial conditions of the universe: A sign of the sine mode

In the standard big bang model the universe starts in a radiation dominated era, where the gravitational perturbations are described by second order differential equations, which will generally have two orthogonal set of solutions. One is the so called {\it growing(cosine)} mode and the other is the {\it decaying(sine)} mode, where the nomenclature is derived from their behaviour on super-horizon(sub-horizon) scales. The decaying mode is qualitatively different to the growing mode of adiabatic perturbations as it evolves with time on \emph{super-horizon} scales. The time dependence of this mode on super-horizon scales is analysed in both the synchronous gauge and the Newtonian gauge to understand the true gauge invariant behaviour of these modes. We then explore constraints on the amplitude of this mode on scales between $k \sim 10^{-5}$ Mpc$^{-1}$ and $k \sim 10^{-1}$ Mpc$^{-1}$ using the temperature and polarization anisotropies from the cosmic microwave background, by computing the Fisher information. Binning the primordial power non-parametrically into 100 bins, we find that the decaying modes are constrained at comparable variance as the growing modes on scales smaller than the horizon today using temperature anisotropies. Adding polrisation data makes the decaying mode more constrained. The decaying mode amplitude is thus constrained by $\sim 1/l$ of the growing mode. On super-horizon scales, the growing mode is poorly constrained, while the decaying mode cannot substantially exceed the scale-invariant amplitude. This interpretation differs substantially from the past literature, where the constraints were quoted in gauge-dependent variables, and resulted in illusionary tight super-horizon decaying mode constraints. The results presented here can generally be used to non-parametrically constrain any model of the early universe.

astro-ph.CO↗

arXiv2019

Effect of choice of probability distribution, randomness, and search methods for alignment modeling in sequence-to-sequence text-to-speech synthesis using hard alignment

Sequence-to-sequence text-to-speech (TTS) is dominated by soft-attention-based methods. Recently, hard-attention-based methods have been proposed to prevent fatal alignment errors, but their sampling method of discrete alignment is poorly investigated. This research investigates various combinations of sampling methods and probability distributions for alignment transition modeling in a hard-alignment-based sequence-to-sequence TTS method called SSNT-TTS. We clarify the common sampling methods of discrete variables including greedy search, beam search, and random sampling from a Bernoulli distribution in a more general way. Furthermore, we introduce the binary Concrete distribution to model discrete variables more properly. The results of a listening test shows that deterministic search is more preferable than stochastic search, and the binary Concrete distribution is robust with stochastic search for natural alignment transition.

eess.AS↗

arXiv2019

Latent Suicide Risk Detection on Microblog via Suicide-Oriented Word Embeddings and Layered Attention

Despite detection of suicidal ideation on social media has made great progress in recent years, people's implicitly and anti-real contrarily expressed posts still remain as an obstacle, constraining the detectors to acquire higher satisfactory performance. Enlightened by the hidden "tree holes" phenomenon on microblog, where people at suicide risk tend to disclose their inner real feelings and thoughts to the microblog space whose authors have committed suicide, we explore the use of tree holes to enhance microblog-based suicide risk detection from the following two perspectives. (1) We build suicide-oriented word embeddings based on tree hole contents to strength the sensibility of suicide-related lexicons and context based on tree hole contents. (2) A two-layered attention mechanism is deployed to grasp intermittently changing points from individual's open blog streams, revealing one's inner emotional world more or less. Our experimental results show that with suicide-oriented word embeddings and attention, microblog-based suicide risk detection can achieve over 91\% accuracy. A large-scale well-labelled suicide data set is also reported in the paper.

cs.CL↗

arXiv2019

Few-shot Object Detection via Feature Reweighting

Conventional training of a deep CNN based object detector demands a large number of bounding box annotations, which may be unavailable for rare categories. In this work we develop a few-shot object detector that can learn to detect novel objects from only a few annotated examples. Our proposed model leverages fully labeled base classes and quickly adapts to novel classes, using a meta feature learner and a reweighting module within a one-stage detection architecture. The feature learner extracts meta features that are generalizable to detect novel object classes, using training data from base classes with sufficient samples. The reweighting module transforms a few support examples from the novel classes to a global vector that indicates the importance or relevance of meta features for detecting the corresponding objects. These two modules, together with a detection prediction module, are trained end-to-end based on an episodic few-shot learning scheme and a carefully designed loss function. Through extensive experiments we demonstrate that our model outperforms well-established baselines by a large margin for few-shot object detection, on multiple datasets and settings. We also present analysis on various aspects of our proposed model, aiming to provide some inspiration for future few-shot detection works.

cs.CV↗

arXiv2019

One-shot entanglement distillation beyond local operations and classical communication

We study the task of entanglement distillation in the one-shot setting under different classes of quantum operations which extend the set of local operations and classical communication (LOCC). Establishing a general formalism which allows for a straightforward comparison of their exact achievable performance, we relate the fidelity of distillation under these classes of operations with a family of entanglement monotones and the rates of distillation with a class of smoothed entropic quantities based on the hypothesis testing relative entropy. We then characterise exactly the one-shot distillable entanglement of several classes of quantum states and reveal many simplifications in their manipulation. We show in particular that the $\varepsilon$-error one-shot distillable entanglement of any pure state is the same under all sets of operations ranging from one-way LOCC to separability-preserving operations or operations preserving the set of states with positive partial transpose, and can be computed exactly as a quadratically constrained linear program. We establish similar operational equivalences in the distillation of isotropic and maximally correlated states, reducing the computation of the relevant quantities to linear or semidefinite programs. We also show that all considered sets of operations achieve the same performance in environment-assisted entanglement distillation from any state.

quant-ph↗

arXiv2019

Large and moderate deviations for a $\mathbb{R}^d$-valued branching random walk with a random environment in time

We consider a $\mathbb{R}^d$-valued branching random walk with a stationary and ergodic environment $ξ=(ξ_n)$ indexed by time $n\in\mathbb{N}$. Let $Z_n$ be the counting measure of particles of generation $n$. With the help of the uniform convergence of martingale and the multifractal analysis, we establish a large deviation result for the measures $Z_n$ as well as a moderate deviation principle.

math.PR↗