arXiv subjects

Yu Wang

Publications and source records attributed to Yu Wang.

At least 1,243 records · Page 69Linked to original sources

arXiv2019

Fuzzy Logic based Autonomous Parking Systems -- Part IV: A Multiple-Model Adaptive Neural-Fuzzy Controller

In this paper, a Multiple Models Adaptive Fuzzy Logic Controller (MM-AFLC) with Neural Network Identification is designed to control the unmanned vehicle in Intelligent Autonomous Parking System. The objective is to achieve robust control while maintaining a low implementation cost. The proposed controller design incorporates the following control theorems -- non-linear system identification using neural network, fuzzy logic control, adaptive control as well as multiple models adaptation. Such integration ensures superior performance compared to previous work. The generalized controller can be applied to different systems without prior knowledge of the actual plant model. In the intelligent autonomous parking system, the proposed controller can be used for both vehicle speed control and steering wheel turning. With a multiple model adaptive fuzzy logic controller, robustness can be also assured under various operating environments regardless of unpredictable disturbances. Last but not least, comparative experiments have also demonstrated that systems equipped with the new controller are able to achieve faster and smoother convergence.

eess.SY↗

arXiv2019

Ferromagnetic van der Waals compound MnSb$_{1.8}$Bi$_{0.2}$Te$_4$

The intersection of topology and magnetism represents a new playground to discover novel quantum phenomena and device concepts. In this work, we show that a van der Waals compound MnSb$_{1.8}$Bi$_{0.2}$Te$_4$ exhibits a ferromagnetic ground state with a Curie temperature of 26 K, in contrast to the antiferromagnetic order previously found for other members of the Mn(Sb, Bi)$_2$Te$_4$ family. We employ magneto-transport, bulk magnetization and neutron scattering studies to illustrate the magnetic and electrical properties of MnSb$_{1.8}$Bi$_{0.2}$Te$_4$ and report on the observation of an unusual anomalous Hall effect. Our results are an important step in the synthesis and understanding of ferromagnetic topological insulators.

cond-mat.mes-hall↗

arXiv2019

Variational Resampling Based Assessment of Deep Neural Networks under Distribution Shift

A novel variational inference based resampling framework is proposed to evaluate the robustness and generalization capability of deep learning models with respect to distribution shift. We use Auto Encoding Variational Bayes to find a latent representation of the data, on which a Variational Gaussian Mixture Model is applied to deliberately create distribution shift by dividing the dataset into different clusters. Wasserstein distance is used to characterize the extent of distribution shift between the generated data splits. We compare several popular Convolutional Neural Network (CNN) architectures and Bayesian CNN models for image classification on the Fashion-MNIST dataset, to assess their robustness and generalization behavior under the deliberately created distribution shift, as well as under random Cross Validation. Our method of creating artificial domain splits of a single dataset can also be used to establish novel model selection criteria and assessment tools in machine learning, as well as benchmark methods for domain adaptation and domain generalization approaches.

cs.LG↗

arXiv2019

A Debiased MDI Feature Importance Measure for Random Forests

Tree ensembles such as Random Forests have achieved impressive empirical success across a wide variety of applications. To understand how these models make predictions, people routinely turn to feature importance measures calculated from tree ensembles. It has long been known that Mean Decrease Impurity (MDI), one of the most widely used measures of feature importance, incorrectly assigns high importance to noisy features, leading to systematic bias in feature selection. In this paper, we address the feature selection bias of MDI from both theoretical and methodological perspectives. Based on the original definition of MDI by Breiman et al. for a single tree, we derive a tight non-asymptotic bound on the expected bias of MDI importance of noisy features, showing that deep trees have higher (expected) feature selection bias than shallow ones. However, it is not clear how to reduce the bias of MDI using its existing analytical expression. We derive a new analytical expression for MDI, and based on this new expression, we are able to propose a debiased MDI feature importance measure using out-of-bag samples, called MDI-oob. For both the simulated data and a genomic ChIP dataset, MDI-oob achieves state-of-the-art performance in feature selection from Random Forests for both deep and shallow trees.

stat.ML↗

arXiv2019

Giant room temperature anomalous Hall effect and magnetically tuned topology in the ferromagnetic Weyl semimetal Co2MnAl

Weyl semimetals (WSM) have been extensively studied due to their exotic properties such as topological surface states and anomalous transport phenomena. Their band structure topology is usually predetermined by material parameters and can hardly be manipulated once the material is formed. Their unique transport properties appear usually at very low temperature, which sets challenges for practical device applications. In this work, we demonstrate a way to modify the band topology via a weak magnetic field in a ferromagnetic topological semimetal, Co2MnAl, at room temperature. We observe a tunable, giant anomalous Hall effect, which is induced by the transition between Weyl points and nodal rings as rotating the magnetization axis. The anomalous Hall conductivity is as large as that of a 3D quantum anomalous Hall effect (QAHE), with the Hall angle reaching a record value (21%) at the room temperature among magnetic conductors. Furthermore, we propose a material recipe to generate the giant anomalous Hall effect by gaping nodal rings without requiring the existence of Weyl points. Our work reveals an ideal intrinsically magnetic platform to explore the interplay between magnetic dynamics and topological physics for the development of a new generation of spintronic devices.

cond-mat.mtrl-sci↗

arXiv2019

Transport evidence of triply degenerate nodal semimetal YRh6Ge4

We have investigated magnetotransport properties of YRh6Ge4, which was recently predicted to be a triply degenerate nodal semimetal. We find it exhibits remarkable signatures of a chiral anomaly, manifested by large negative longitudinal magnetoresistance, quadratic field dependence of magnetoconductance and planar Hall effect. Furthermore, we have also observed Shubnikov-de Haas (SdH) quantum oscillations in the magnetoresistivity measurements on this material. The analyses of the SdH data reveal two point-like Fermi surfaces and these pockets are found to host nearly massless fermions. The small size of these Fermi pockets is in a good agreement with the theoretical prediction that the triply degenerate point in YRh6Ge4 is much closer to the Fermi level than previously demonstrated triply degenerate nodal semimetals such as MoP and WC. These results suggest YRh6Ge4 may serve as a model system to probe exotic properties of three-component fermions and understand their underlying physics.

physics.app-ph↗

arXiv2019

Unified Language Model Pre-training for Natural Language Understanding and Generation

This paper presents a new Unified pre-trained Language Model (UniLM) that can be fine-tuned for both natural language understanding and generation tasks. The model is pre-trained using three types of language modeling tasks: unidirectional, bidirectional, and sequence-to-sequence prediction. The unified modeling is achieved by employing a shared Transformer network and utilizing specific self-attention masks to control what context the prediction conditions on. UniLM compares favorably with BERT on the GLUE benchmark, and the SQuAD 2.0 and CoQA question answering tasks. Moreover, UniLM achieves new state-of-the-art results on five natural language generation datasets, including improving the CNN/DailyMail abstractive summarization ROUGE-L to 40.51 (2.04 absolute improvement), the Gigaword abstractive summarization ROUGE-L to 35.75 (0.86 absolute improvement), the CoQA generative question answering F1 score to 82.5 (37.1 absolute improvement), the SQuAD question generation BLEU-4 to 22.12 (3.75 absolute improvement), and the DSTC7 document-grounded dialog response generation NIST-4 to 2.67 (human performance is 2.65). The code and pre-trained models are available at https://github.com/microsoft/unilm.

cs.CL↗

arXiv2019

Hidden Talents of the Variational Autoencoder

Variational autoencoders (VAE) represent a popular, flexible form of deep generative model that can be stochastically fit to samples from a given random process using an information-theoretic variational bound on the true underlying distribution. Once so-obtained, the model can be putatively used to generate new samples from this distribution, or to provide a low-dimensional latent representation of existing samples. While quite effective in numerous application domains, certain important mechanisms which govern the behavior of the VAE are obfuscated by the intractable integrals and resulting stochastic approximations involved. Moreover, as a highly non-convex model, it remains unclear exactly how minima of the underlying energy relate to original design purposes. We attempt to better quantify these issues by analyzing a series of tractable special cases of increasing complexity. In doing so, we unveil interesting connections with more traditional dimensionality reduction models, as well as an intrinsic yet underappreciated propensity for robustly dismissing sparse outliers when estimating latent manifolds. With respect to the latter, we demonstrate that the VAE can be viewed as the natural evolution of recent robust PCA models, capable of learning nonlinear manifolds of unknown dimension obscured by gross corruptions.

cs.LG↗

arXiv2019

Boron isotope effect on the thermal conductivity of boron arsenide single crystals

Boron arsenide (BAs) with a zinc blende structure has recently been discovered to exhibit unusual and ultrahigh thermal conductivity (k), providing a new outlook for research on BAs and other high thermal conductivity materials. Technology for BAs crystal growth has been continuously improving, however, the influence of boron isotopes, pure or mixed, on the thermal conductivity in BAs is still not completely clear. Here we report detailed studies on the growth of single crystals of BAs with different isotopic ratios and demonstrate that the k of isotopically pure BAs is at least 10% higher than that of BAs grown from natural B. Raman spectroscopy characterization shows differences in scattering among various BAs samples. The presented results will be helpful in guiding further studies on the influence of isotopes on optimizing k in BAs.

cond-mat.mtrl-sci↗

arXiv2019

Non-native Speaker Verification for Spoken Language Assessment

Automatic spoken language assessment systems are becoming more popular in order to handle increasing interests in second language learning. One challenge for these systems is to detect malpractice. Malpractice can take a range of forms, this paper focuses on detecting when a candidate attempts to impersonate another in a speaking test. This form of malpractice is closely related to speaker verification, but applied in the specific domain of spoken language assessment. Advanced speaker verification systems, which leverage deep-learning approaches to extract speaker representations, have been successfully applied to a range of native speaker verification tasks. These systems are explored for non-native spoken English data in this paper. The data used for speaker enrolment and verification is mainly taken from the BULATS test, which assesses English language skills for business. Performance of systems trained on relatively limited amounts of BULATS data, and standard large speaker verification corpora, is compared. Experimental results on large-scale test sets with millions of trials show that the best performance is achieved by adapting the imported model to non-native data. Breakdown of impostor trials across different first languages (L1s) and grades is analysed, which shows that inter-L1 impostors are more challenging for speaker verification systems.

eess.AS↗

arXiv2019

Single Training Dimension Selection for Word Embedding with PCA

In this paper, we present a fast and reliable method based on PCA to select the number of dimensions for word embeddings. First, we train one embedding with a generous upper bound (e.g. 1,000) of dimensions. Then we transform the embeddings using PCA and incrementally remove the lesser dimensions one at a time while recording the embeddings' performance on language tasks. Lastly, we select the number of dimensions while balancing model size and accuracy. Experiments using various datasets and language tasks demonstrate that we are able to train 10 times fewer sets of embeddings while retaining optimal performance. Researchers interested in training the best-performing embeddings for downstream tasks, such as sentiment analysis, question answering and hypernym extraction, as well as those interested in embedding compression should find the method helpful.

cs.CL↗

arXiv2019

On Heavy-user Bias in A/B Testing

On-line experimentation (also known as A/B testing) has become an integral part of software development. To timely incorporate user feedback and continuously improve products, many software companies have adopted the culture of agile deployment, requiring online experiments to be conducted and concluded on limited sets of users for a short period. While conceptually efficient, the result observed during the experiment duration can deviate from what is seen after the feature deployment, which makes the A/B test result biased. In this paper, we provide theoretical analysis to show that heavy-users can contribute significantly to the bias, and propose a re-sampling estimator for bias adjustment.

stat.AP↗

arXiv2019

Statistical Verification of Hyperproperties for Cyber-Physical System

Many important properties of cyber-physical systems (CPS) are defined upon the relationship between multiple executions simultaneously in continuous time. Examples include probabilistic fairness and sensitivity to modeling errors (i.e., parameters changes) for real-valued signals. These requirements can only be specified by hyperproperties. In this work, we focus on verifying probabilistic hyperproperties for CPS. To cover a wide range of modeling formalisms, we first propose a general model of probabilistic uncertain systems (PUSs) that unify commonly studied CPS models such as continuous-time Markov chains (CTMCs) and probabilistically parametrized Hybrid I/O Automata. To formally specify hyperproperties, we propose a new temporal logic, hyper probabilistic signal temporal logic (HyperPSTL) that serves as a hyper and probabilistic version of the conventional signal temporal logic (STL). Considering complexity of real-world systems that can be captured as PUSs, we adopt a statistical model checking (SMC) approach for their verification. We develop a new SMC technique based on the direct computation of the significance levels of statistical assertions for HyperPSTL specifications, which requires no a priori knowledge on the indifference margin. Then, we introduce SMC algorithms for HyperPSTL specifications on the joint probabilistic distribution of multiple paths, as well as specifications with nested probabilistic operators quantifying different paths, which cannot be handled by existing SMC algorithms. Finally, we show the effectiveness of our SMC algorithms on CPS benchmarks with varying levels of complexity, including the Toyota Powertrain Control~System.

cs.LO↗

arXiv2019

Optimal Dynamic Multi-Resource Management in Earth Observation Oriented Space Information Networks

Space information network (SIN) is an innovative networking architecture to achieve near-real-time mass data observation, processing and transmission over the globe. In the SIN environment, it is essential to coordinate multi-dimensional heterogeneous resources (i.e., observation resource, computation resource and transmission resource) to improve network performance. However, the time varying property of both the observation resource and transmission resource is not fully exploited in existing studies. Dynamic resource management according to instantaneous channel conditions has a potential to enhance network performance. To this end, in this paper, we study the multi-resource dynamic management problem, considering stochastic observation and transmission channel conditions in SINs. Specifically, we develop an aggregate optimization framework for observation scheduling, compression ratio selection and transmission scheduling, and formulate a flow optimization problem based on extended time expanded graph (ETEG) to maximize the sum network utility. Then, we equivalently transform the flow optimization problem on ETEG as a queue stability-related stochastic optimization problem. An online algorithm is proposed to solve the problem in a slot-by-slot manner by exploiting the Lyapunov optimization technique. Performance analysis shows that the proposed algorithm achieves close-to-optimal network utility while guaranteeing bounded queue occupancy. Extensive simulation results further validate the efficiency of the proposed algorithm and evaluate the impacts of various network parameters on the algorithm performance.

eess.SY↗

arXiv2019

DNNVM : End-to-End Compiler Leveraging Heterogeneous Optimizations on FPGA-based CNN Accelerators

The convolutional neural network (CNN) has become a state-of-the-art method for several artificial intelligence domains in recent years. The increasingly complex CNN models are both computation-bound and I/O-bound. FPGA-based accelerators driven by custom instruction set architecture (ISA) achieve a balance between generality and efficiency, but there is much on them left to be optimized. We propose the full-stack compiler DNNVM, which is an integration of optimizers for graphs, loops and data layouts, and an assembler, a runtime supporter and a validation environment. The DNNVM works in the context of deep learning frameworks and transforms CNN models into the directed acyclic graph: XGraph. Based on XGraph, we transform the optimization challenges for both the data layout and pipeline into graph-level problems. DNNVM enumerates all potentially profitable fusion opportunities by a heuristic subgraph isomorphism algorithm to leverage pipeline and data layout optimizations, and searches for the best choice of execution strategies of the whole computing graph. On the Xilinx ZU2 @330 MHz and ZU9 @330 MHz, we achieve equivalently state-of-the-art performance on our benchmarks by naïve implementations without optimizations, and the throughput is further improved up to 1.26x by leveraging heterogeneous optimizations in DNNVM. Finally, with ZU9 @330 MHz, we achieve state-of-the-art performance for VGG and ResNet50. We achieve a throughput of 2.82 TOPs/s and an energy efficiency of 123.7 GOPs/s/W for VGG. Additionally, we achieve 1.38 TOPs/s for ResNet50 and 1.41 TOPs/s for GoogleNet.

cs.DC↗

arXiv2019

Slowing-down reduction and Possible Reversal Trend of Tropospheric NO2 over China during 2016 to 2019

Atmospheric nitrogen dioxide (NO2) over China at national level has been kept reducing since 2011 as seen from both satellite observations, ground-based measurements and bottom-up emission inventory (Liu et al., 2016; Irie et al., 2016; Krotkov et al., 2016; Foy et al., 2016; Liu et al 2017). These studies used data before 2015-2016. After 2016, however, a significant slowing-down of the reduction trend and/or even a reversal trend were found in numerous provinces, particularly in those with heavy NO2 level, based on satellite observations. Error analysis on satellite data excluded cloud contamination, instrument anomalies from the main reasons of this change. Ground-based measurements show strong positive correlations with satellite observations and similar patterns of year-to-year changes of NO2 in 2018 and 2019 winter time. The temporal variations of Satellite NO2 over China are believed largely determined by surface emission from power plant and transportation. The reason for the recent change from emission perspective cannot be traced down since the national emission inventory was not updated since 2015. We therefore call on immediate attentions from both scientist community and policy makers to this phenomenon. Further efforts should be made to understand the reasons causing this change and to make associated air pollution controlling actions.

physics.ao-ph↗

arXiv2019

Approximately Optimal Subset Selection for Statistical Design and Modelling

We study the problem of optimal subset selection from a set of correlated random variables. In particular, we consider the associated combinatorial optimization problem of maximizing the determinant of a symmetric positive definite matrix that characterizes the chosen subset. This problem arises in many domains, such as experimental designs, regression modeling, and environmental statistics. We establish an efficient polynomial-time algorithm using Determinantal Point Process for approximating the optimal solution to the problem. We demonstrate the advantages of our methods by presenting computational results for both synthetic and real data sets.

stat.CO↗

arXiv2019

ESNet: An Efficient Symmetric Network for Real-time Semantic Segmentation

The recent years have witnessed great advances for semantic segmentation using deep convolutional neural networks (DCNNs). However, a large number of convolutional layers and feature channels lead to semantic segmentation as a computationally heavy task, which is disadvantage to the scenario with limited resources. In this paper, we design an efficient symmetric network, called (ESNet), to address this problem. The whole network has nearly symmetric architecture, which is mainly composed of a series of factorized convolution unit (FCU) and its parallel counterparts (PFCU). On one hand, the FCU adopts a widely-used 1D factorized convolution in residual layers. On the other hand, the parallel version employs a transform-split-transform-merge strategy in the designment of residual module, where the split branch adopts dilated convolutions with different rate to enlarge receptive field. Our model has nearly 1.6M parameters, and is able to be performed over 62 FPS on a single GTX 1080Ti GPU. The experiments demonstrate that our approach achieves state-of-the-art results in terms of speed and accuracy trade-off for real-time semantic segmentation on CityScapes dataset.

cs.CV↗