arXiv subjects

Jun Huang

Publications and source records attributed to Jun Huang.

At least 181 records · Page 10Linked to original sources

arXiv2020

Enforced freedom: electric-field-induced declustering of ionic-liquid ions in the double layer

Whereas the majority of ions in the bulk of a solvent-free ionic liquid is bound into clusters, this is expected to change in the electrical double layer (EDL), in which the resulting electric field 'prefers' to interact with electrical monopoles-free, unclustered ions. The competition between the propensity of ions to stay in a clustered state and the reduction of the energy of ions in electric field in the free state determines the resulting portion of free ions in the EDL. We present a study of this effect, based on the simplest possible mean-field theory. 'Cracking' of ion clusters into individual ions in electric field is accompanied by the change of the dielectric response of ionic liquid which is different in clustered and unclustered states. The predictions of the theory are verified and further explored by specially performed molecular dynamics simulations. A particular finding of the theory is that the differential capacitance vs potential curve displays a bell shape despite low concentration of free charge carriers, because the dielectric response of bound ions reduces the threshold concentration of the bell- to camel-shape transition. Whereas qualitatively these findings make perfect sense, in reality the exact numbers and criteria might be different as the presented simple theory does not take into account overscreening and oscillating charge and electrostatic potential distributions near the electrode. This is why testing the theory with computer simulations is essential, but the latter basically reproduce the qualitative conclusions of the theory.

physics.chem-ph↗

arXiv2020

Attack-Aware Data Timestamping in Low-Power Synchronization-Free LoRaWAN

Low-power wide-area network technologies such as LoRaWAN are promising for collecting low-rate monitoring data from geographically distributed sensors, in which timestamping the sensor data is a critical system function. This paper considers a synchronization-free approach to timestamping LoRaWAN uplink data based on signal arrival time at the gateway, which well matches LoRaWAN's one-hop star topology and releases bandwidth from transmitting timestamps and synchronizing end devices' clocks at all times. However, we show that this approach is susceptible to a {\em frame delay attack} consisting of malicious frame collision and delayed replay. Real experiments show that the attack can affect the end devices in large areas up to about $50,000\,\text{m}^2$. In a broader sense, the attack threatens any system functions requiring timely deliveries of LoRaWAN frames. To address this threat, we propose a $\mathsf{LoRaTS}$ gateway design that integrates a commodity LoRaWAN gateway and a low-power software-defined radio receiver to track the inherent frequency biases of the end devices. Based on an analytic model of LoRa's chirp spread spectrum modulation, we develop signal processing algorithms to estimate the frequency biases with high accuracy beyond that achieved by LoRa's default demodulation. The accurate frequency bias tracking capability enables the detection of the attack that introduces additional frequency biases. Extensive experiments show the effectiveness of our approach.

cs.NI↗

arXiv2020

Plasma Surface Metallurgy of Materials Based on Double Glow Discharge Phenomenon

Plasma Surface Metallurgy/Alloying is a kind of surface metallurgy/alloying to employ low temperature plasma produced by glow discharge to diffuse alloying elements into the surface of substrate material to form an alloy layer. The first plasma surface metallurgy technology is plasma nitriding invented by German scientist Dr. Bernard Berghuas in 1930. He was the first person to apply glow discharge to realize the surface alloying. In order to break the limitation of plasma nitriding technology, which can only be applied to a few non-metallic gaseous elements such as nitrogen, carbon, sulfur, the "Double Glow Discharge Phenomenon"was found in 1978. Based on this phenomenon the "Double Glow Plasma Surface Metallurgy Technology", also known as the "Xu-Tec Process" was invented in 1980. It can utilize any chemical elements in the periodic table including solid metallic, gas non-metallic elements and their combination to realize plasma surface alloying, hence greatly expanded the field of surface alloying. Countless surface alloys with high hardness, wear resistance and corrosion resistance, such as high speed steels, nickel base alloys and burn resistant alloys have been produced on the surfaces of a variety of materials. This technology may greatly improve the surface properties of metal materials, comprehensively improve the quality of mechanical products, save a lot of precious alloy elements for human beings. Based on the plasma nitriding technology, the Xu-Tec Process has opened up a new material engineering field of "Plasma Surface Metallurgy". This Review Article briefly presents the history of glow discharge and surface alloying, double glow discharge phenomenon, basic principle and current status of Double Glow Plasma Surface Metallurgy/Alloying. Industrial applications, advantages and future potential of the Xu-Tec process are also presented.

physics.app-ph↗

arXiv2020

A Series of Plasma Innovation Technologies by Double Glow Discharge Phenomenon

In order to break the limitation of plasma nitriding technology,which can be applied to a few nonmetallic gaseous elements, the "Double Glow Discharge Phenomenon" was found and then invented the "Double Glow Plasma Surface Metallurgy Technology". This double glow plasma surface metallurgy technology can use any element in the periodic table of chemical elements for surface alloying of metal materials. Countless surface alloys with special physical and chemical properties have been produced on the surfaces of conductive materials.By using double glow discharge phenomenon,a series of new plasma technologies,such as the double glow plasma graphene technology, double glow plasma brazing technology,double glow plasma sintering technology, double glow plasma nanotechnology,double glow plasma cleaning technology, double glow plasma carburizing without hydrogen and so on, have been invented.A very simple phenomenon of double glow discharge can generate about 10 plasma innovation technologies, which fully shows that there is still a lot of innovation space on the basis of classical physics.This paper briefly introduces the basic principles,functions and characteristics of each technology. The application prospects and development directions of plasma in metal materials and machinery manufacturing industry will also be discussed.

physics.app-ph↗

arXiv2020

SwapText: Image Based Texts Transfer in Scenes

Swapping text in scene images while preserving original fonts, colors, sizes and background textures is a challenging task due to the complex interplay between different factors. In this work, we present SwapText, a three-stage framework to transfer texts across scene images. First, a novel text swapping network is proposed to replace text labels only in the foreground image. Second, a background completion network is learned to reconstruct background images. Finally, the generated foreground image and background image are used to generate the word image by the fusion network. Using the proposing framework, we can manipulate the texts of the input images even with severe geometric distortion. Qualitative and quantitative results are presented on several scene text datasets, including regular and irregular text datasets. We conducted extensive experiments to prove the usefulness of our method such as image based text translation, text image synthesis, etc.

cs.CV↗

arXiv2019

Internet of Autonomous Vehicles: Architecture, Features, and Socio-Technological Challenges

Mobility is the backbone of urban life and a vital economic factor in the development of the world. Rapid urbanization and the growth of mega-cities is bringing dramatic changes in the capabilities of vehicles. Innovative solutions like autonomy, electrification, and connectivity are on the horizon. How, then, we can provide ubiquitous connectivity to the legacy and autonomous vehicles? This paper seeks to answer this question by combining recent leaps of innovation in network virtualization with remarkable feats of wireless communications. To do so, this paper proposes a novel paradigm called the Internet of autonomous vehicles (IoAV). We begin painting the picture of IoAV by discussing the salient features, and applications of IoAV which is followed by a detailed discussion on the key enabling technologies. Next, we describe the proposed layered architecture of IoAV and uncover some critical functions of each layer. This is followed by the performance evaluation of IoAV which shows the significant advantage of the proposed architecture in terms of transmission time and energy consumption. Finally, to best capture the benefits of IoAV, we enumerate some social and technological challenges and explain how some unresolved issues can disrupt the widespread use of autonomous vehicles in the future.

cs.NI↗

arXiv2019

Artificial Intelligence for Prosthetics - challenge solutions

In the NeurIPS 2018 Artificial Intelligence for Prosthetics challenge, participants were tasked with building a controller for a musculoskeletal model with a goal of matching a given time-varying velocity vector. Top participants were invited to describe their algorithms. In this work, we describe the challenge and present thirteen solutions that used deep reinforcement learning approaches. Many solutions use similar relaxations and heuristics, such as reward shaping, frame skipping, discretization of the action space, symmetry, and policy blending. However, each team implemented different modifications of the known algorithms by, for example, dividing the task into subtasks, learning low-level control, or by incorporating expert knowledge and using imitation learning.

cs.LG↗

arXiv2018

Learning to Selectively Transfer: Reinforced Transfer Learning for Deep Text Matching

Deep text matching approaches have been widely studied for many applications including question answering and information retrieval systems. To deal with a domain that has insufficient labeled data, these approaches can be used in a Transfer Learning (TL) setting to leverage labeled data from a resource-rich source domain. To achieve better performance, source domain data selection is essential in this process to prevent the "negative transfer" problem. However, the emerging deep transfer models do not fit well with most existing data selection methods, because the data selection policy and the transfer learning model are not jointly trained, leading to sub-optimal training efficiency. In this paper, we propose a novel reinforced data selector to select high-quality source domain data to help the TL model. Specifically, the data selector "acts" on the source domain data to find a subset for optimization of the TL model, and the performance of the TL model can provide "rewards" in turn to update the selector. We build the reinforced data selector based on the actor-critic framework and integrate it to a DNN based transfer learning model, resulting in a Reinforced Transfer Learning (RTL) method. We perform a thorough experimental evaluation on two major tasks for text matching, namely, paraphrase identification and natural language inference. Experimental results show the proposed RTL can significantly improve the performance of the TL model. We further investigate different settings of states, rewards, and policy optimization methods to examine the robustness of our method. Last, we conduct a case study on the selected data and find our method is able to select source domain data whose Wasserstein distance is close to the target domain data. This is reasonable and intuitive as such source domain data can provide more transferability power to the model.

cs.IR↗

arXiv2018

Review Helpfulness Prediction with Embedding-Gated CNN

Product reviews, in the form of texts dominantly, significantly help consumers finalize their purchasing decisions. Thus, it is important for e-commerce companies to predict review helpfulness to present and recommend reviews in a more informative manner. In this work, we introduce a convolutional neural network model that is able to extract abstract features from multi-granularity representations. Inspired by the fact that different words contribute to the meaning of a sentence differently, we consider to learn word-level embedding-gates for all the representations. Furthermore, as it is common that some product domains/categories have rich user reviews, other domains not. To help domains with less sufficient data, we integrate our model into a cross-domain relationship learning framework for effectively transferring knowledge from other domains. Extensive experiments show that our model yields better performance than the existing methods.

cs.CL↗

arXiv2018

Transfer Learning for Context-Aware Question Matching in Information-seeking Conversations in E-commerce

Building multi-turn information-seeking conversation systems is an important and challenging research topic. Although several advanced neural text matching models have been proposed for this task, they are generally not efficient for industrial applications. Furthermore, they rely on a large amount of labeled data, which may not be available in real-world applications. To alleviate these problems, we study transfer learning for multi-turn information seeking conversations in this paper. We first propose an efficient and effective multi-turn conversation model based on convolutional neural networks. After that, we extend our model to adapt the knowledge learned from a resource-rich domain to enhance the performance. Finally, we deployed our model in an industrial chatbot called AliMe Assist (https://consumerservice.taobao.com/online-help) and observed a significant improvement over the existing online model.

cs.CL↗

arXiv2018

Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems

Intelligent personal assistant systems with either text-based or voice-based conversational interfaces are becoming increasingly popular around the world. Retrieval-based conversation models have the advantages of returning fluent and informative responses. Most existing studies in this area are on open domain "chit-chat" conversations or task / transaction oriented conversations. More research is needed for information-seeking conversations. There is also a lack of modeling external knowledge beyond the dialog utterances among current conversational models. In this paper, we propose a learning framework on the top of deep neural matching networks that leverages external knowledge for response ranking in information-seeking conversation systems. We incorporate external knowledge into deep neural models with pseudo-relevance feedback and QA correspondence knowledge distillation. Extensive experiments with three information-seeking conversation data sets including both open benchmarks and commercial data show that, our methods outperform various baseline methods including several deep text matching models and the state-of-the-art method on response selection in multi-turn conversations. We also perform analysis over different response types, model variations and ranking examples. Our models and research findings provide new insights on how to utilize external knowledge with deep neural models for response selection and have implications for the design of the next generation of information-seeking conversation systems.

cs.IR↗

arXiv2018

Magnetically Localizing Heat or Enhancing Equilibration in a Quasi 1D Magnetic Fluid

Using two different configurations of temperature and magnetic field gradients, we observed that, in a quasi one-dimensional magnetic fluid, magnetic force either reduces the temperature difference across the sample when the two gradients are parallel to each other (PL), or increase the temperature difference when the two gradients are antiparallel (AP), where the single convection roll in zero field was replaced by two localized flows at the two ends of the sample cell. This flow structure stops the heat flow of approaching to thermal equilibrium in the system, causing the temperature at hot side of the sample cell getting hotter and cold side becoming colder. None of these phenomena can be described by the existing theories of magnetically-induced instabilities. The underlying physics for observed results for AP configuration has been proposed as the mechanism to drive a new type of heat engines that has much higher efficiency than Carnot engines and has no pollution to the environment, our results point to the potential feasibility of this proposed mechanism.

cond-mat.soft↗

arXiv2018

AliMe Assist: An Intelligent Assistant for Creating an Innovative E-commerce Experience

We present AliMe Assist, an intelligent assistant designed for creating an innovative online shopping experience in E-commerce. Based on question answering (QA), AliMe Assist offers assistance service, customer service, and chatting service. It is able to take voice and text input, incorporate context to QA, and support multi-round interaction. Currently, it serves millions of customer questions per day and is able to address 85% of them. In this paper, we demonstrate the system, present the underlying techniques, and share our experience in dealing with real-world QA in the E-commerce field.

cs.CL↗

arXiv2017

Modelling Domain Relationships for Transfer Learning on Retrieval-based Question Answering Systems in E-commerce

In this paper, we study transfer learning for the PI and NLI problems, aiming to propose a general framework, which can effectively and efficiently adapt the shared knowledge learned from a resource-rich source domain to a resource- poor target domain. Specifically, since most existing transfer learning methods only focus on learning a shared feature space across domains while ignoring the relationship between the source and target domains, we propose to simultaneously learn shared representations and domain relationships in a unified framework. Furthermore, we propose an efficient and effective hybrid model by combining a sentence encoding- based method and a sentence interaction-based method as our base model. Extensive experiments on both paraphrase identification and natural language inference demonstrate that our base model is efficient and has promising performance compared to the competing models, and our transfer learning method can help to significantly boost the performance. Further analysis shows that the inter-domain and intra-domain relationship captured by our model are insightful. Last but not least, we deploy our transfer learning model for PI into our online chatbot system, which can bring in significant improvements over our existing system. Finally, we launch our new system on the chatbot platform Eva in our E-commerce site AliExpress.

cs.CL↗

arXiv2017

PS-DBSCAN: An Efficient Parallel DBSCAN Algorithm Based on Platform Of AI (PAI)

We present PS-DBSCAN, a communication efficient parallel DBSCAN algorithm that combines the disjoint-set data structure and Parameter Server framework in Platform of AI (PAI). Since data points within the same cluster may be distributed over different workers which result in several disjoint-sets, merging them incurs large communication costs. In our algorithm, we employ a fast global union approach to union the disjoint-sets to alleviate the communication burden. Experiments over the datasets of different scales demonstrate that PS-DBSCAN outperforms the PDSDBSCAN with 2-10 times speedup on communication efficiency. We have released our PS-DBSCAN in an algorithm platform called Platform of AI (PAI - https://pai.base.shuju.aliyun.com/) in Alibaba Cloud. We have also demonstrated how to use the method in PAI.

cs.DB↗

arXiv2017

Application-Layer Clock Synchronization for Wearables Using Skin Electric Potentials Induced by Powerline Radiation

Design of clock synchronization for networked nodes faces a fundamental trade-off between synchronization accuracy and universality for heterogeneous platforms, because a high synchronization accuracy generally requires platform-dependent hardware-level network packet timestamping. This paper presents TouchSync, a new indoor clock synchronization approach for wearables that achieves millisecond accuracy while preserving universality in that it uses standard system calls only, such as reading system clock, sampling sensors, and sending/receiving network messages. The design of TouchSync is driven by a key finding from our extensive measurements that the skin electric potentials (SEPs) induced by powerline radiation are salient, periodic, and synchronous on a same wearer and even across different wearers. TouchSync integrates the SEP signal into the universal principle of Network Time Protocol and solves an integer ambiguity problem by fusing the ambiguous results in multiple synchronization rounds to conclude an accurate clock offset between two synchronizing wearables. With our shared code, TouchSync can be readily integrated into any wearable applications. Extensive evaluation based on our Arduino and TinyOS implementations shows that TouchSync's synchronization errors are below 3 and 7 milliseconds on the same wearer and between two wearers 10 kilometers apart, respectively.

cs.DC↗

arXiv2016

Performance degradation of superlattice MOSFETs due to scattering in the contacts

Ideal, completely coherent quantum transport calculations had predicted that superlattice MOSFETs may offer steep subthreshold swing performance below 60mV/dec to around 39mV/dec. However, the high carrier density in the superlattice source suggest that scattering may significantly degrade the ideal device performance. Such effects of electron scattering and decoherence in the contacts of superlattice MOSFETs are examined through a multiscale quantum transport model developed in NEMO5. This model couples NEGF-based quantum ballistic transport in the channel to a quantum mechanical density of states dominated reservoir, which is thermalized through strong scattering with local quasi-Fermi levels determined by drift-diffusion transport. The simulations show that scattering increases the electron transmission in the nominally forbidden minigap therefore degrading the subthreshold swing (S.S.) and the ON/OFF DC current ratio. This degradation varies with both the scattering rate and the length of the scattering dominated regions. Different superlattice MOSFET designs are explored to mitigate the effects of such deleterious scattering. Specifically, shortening the spacer region between the superlattice and the channel from 3.5 nm to 0 nm improves the simulated S.S. from 51mV/dec. to 40mV/dec. I. INTRODUCTION

cond-mat.mes-hall↗

arXiv2016

Discovery of homogeneously dispersed pentacoordinated Al(V) species on the surface of amorphous silica-alumina

The dispersion and coordination of aluminium species on the surface of silica-alumina based materials are essential for controlling their catalytic activity and selectivity. Al(IV) and Al(VI) are two common coordinations of Al species in the silica network and alumina phase, respectively. Al(V) is rare in nature and was found hitherto only in the alumina phase or interfaces containing alumina, a behavior which negatively affects the dispersion, population, and accessibility of Al(V) species on the silica-alumina surface. This constraint has limited the development of silica-alumina based catalysts, particularly because Al(V) had been confirmed to act as a highly active center for acid reactions and single-atom catalysts. Here, we report the direct observation of high population of homogenously dispersed Al(V) species in amorphous silica-alumina in the absence of any bulk alumina phase, by high resolution TEM/EDX and high magnetic-field MAS NMR. Solid-state 27Al multi-quantum MAS NMR experiments prove unambiguously that most of the Al(V) species formed independently from the alumina phase and are accessible on the surface for guest molecules. These species are mainly transferred to Al(VI) species with partial formation of Al(IV) species after adsorption of water. The NMR chemical shifts and their coordination transformation with and without water adsorption are matching that obtained in DFT calculations of the predicted clusters. The discovery presented in this study not only provides fundamental knowledge of the nature of aluminum coordination, but also paves the way for developing highly efficient catalysts.

physics.chem-ph↗