arXiv subjects

Min Yang

Publications and source records attributed to Min Yang.

At least 307 records · Page 17Linked to original sources

arXiv2021

PC-GAIN: Pseudo-label Conditional Generative Adversarial Imputation Networks for Incomplete Data

Datasets with missing values are very common in real world applications. GAIN, a recently proposed deep generative model for missing data imputation, has been proved to outperform many state-of-the-art methods. But GAIN only uses a reconstruction loss in the generator to minimize the imputation error of the non-missing part, ignoring the potential category information which can reflect the relationship between samples. In this paper, we propose a novel unsupervised missing data imputation method named PC-GAIN, which utilizes potential category information to further enhance the imputation power. Specifically, we first propose a pre-training procedure to learn potential category information contained in a subset of low-missing-rate data. Then an auxiliary classifier is determined using the synthetic pseudo-labels. Further, this classifier is incorporated into the generative adversarial framework to help the generator to yield higher quality imputation results. The proposed method can improve the imputation quality of GAIN significantly. Experimental results on various benchmark datasets show that our method is also superior to other baseline approaches. Our code is available at \url{https://github.com/WYu-Feng/pc-gain}.

cs.LG↗

arXiv2021

On the Characterizations of OTFS Modulation over multipath Rapid Fading Channel

Orthogonal time frequency space (OTFS) modulation has been confirmed to provide significant performance advantages against Doppler in high-mobility scenarios. The core feature of OTFS is that the time-variant channel is converted into a non-fading 2D channel in the delay-Doppler (DD) domain so that all symbols experience the same channel gain. In now available literature, the channel is assumed to be quasi-static over an OTFS frame. As for more practical channels, the input-output relation will be time-variant as the environment or medium changes. In this paper, we analyze the characterizations of OTFS modulation over a more general multipath channel, where the signal of each path has experienced a unique rapid fading. First, we derive the explicit input-output relationship of OTFS in the DD domain for the case of ideal pulse and rectangular pulse. It is shown that the rapid fading will produce extra Doppler dispersion without impacting on delay domain. We next demonstrate that OTFS can be interpreted as an efficient time diversity technology that combines space-time encoding and interleaving. Simulation results reveal that OTFS is insensitive to rapid fading and still outperforms orthogonal frequency-division multiplexing (OFDM) in these types of channels.

cs.IT↗

arXiv2020

The Style-Content Duality of Attractiveness: Learning to Write Eye-Catching Headlines via Disentanglement

Eye-catching headlines function as the first device to trigger more clicks, bringing reciprocal effect between producers and viewers. Producers can obtain more traffic and profits, and readers can have access to outstanding articles. When generating attractive headlines, it is important to not only capture the attractive content but also follow an eye-catching written style. In this paper, we propose a Disentanglement-based Attractive Headline Generator (DAHG) that generates headline which captures the attractive content following the attractive style. Concretely, we first devise a disentanglement module to divide the style and content of an attractive prototype headline into latent spaces, with two auxiliary constraints to ensure the two spaces are indeed disentangled. The latent content information is then used to further polish the document representation and help capture the salient part. Finally, the generator takes the polished document as input to generate headline under the guidance of the attractive style. Extensive experiments on the public Kuaibao dataset show that DAHG achieves state-of-the-art performance. Human evaluation also demonstrates that DAHG triggers 22% more clicks than existing models.

cs.CL↗

arXiv2020

Weak Links in Authentication Chains: A Large-scale Analysis of Email Sender Spoofing Attacks

As a fundamental communicative service, email is playing an important role in both individual and corporate communications, which also makes it one of the most frequently attack vectors. An email's authenticity is based on an authentication chain involving multiple protocols, roles and services, the inconsistency among which creates security threats. Thus, it depends on the weakest link of the chain, as any failed part can break the whole chain-based defense. This paper systematically analyzes the transmission of an email and identifies a series of new attacks capable of bypassing SPF, DKIM, DMARC and user-interface protections. In particular, by conducting a "cocktail" joint attack, more realistic emails can be forged to penetrate the celebrated email services, such as Gmail and Outlook. We conduct a large-scale experiment on 30 popular email services and 23 email clients, and find that all of them are vulnerable to certain types of new attacks. We have duly reported the identified vulnerabilities to the related email service providers, and received positive responses from 11 of them, including Gmail, Yahoo, iCloud and Alibaba. Furthermore, we propose key mitigating measures to defend against the new attacks. Therefore, this work is of great value for identifying email spoofing attacks and improving the email ecosystem's overall security.

cs.CR↗

arXiv2020

TDACS: an ABAC and Trust-based Dynamic Access Control Scheme in Hadoop

The era of big data has promoted the vigorous development of many industries, boosting the full potential of holistic data-driven analysis. Hadoop has become the primary choice for mainstream platforms used by stakeholders to process big data. Thereafter, the security of Hadoop platform has arisen tremendous attention worldwide. In this paper, we mainly concentrate on enforcing access control on users to ensure platform security. First, we leverage access proxy integrated with attribute-based access control (ABAC) model to implement front-end authorization, which can fully reflect and cope with the flexible nature of the complex access control process in Hadoop platform, as well as can release back-end resources from complex authorization process through access proxy. Moreover, in order to ensure the fine-granularity of authorization, the access proxy maintains a list composed of trust threshold value provided by each resource according to its importance. The access proxy interacts with the blockchain network to obtain the user's trust evaluation value, which serves as an important basis for dynamic authorization determination. More specifically, blockchain network works together on-chain and off-chain modes. The user's historical behavior data is stored off-chain, and the corresponding hash value is anchored on-chain. Consequently, the user's trust value is evaluated based on his historical behavior stored on the blockchain platform. Meanwhile, the authenticity of user behavior data can be guaranteed, thereby ensuring the reliability of trust assessment results. Our experiment demonstrates that the proposed model can dynamically and flexibly adjust user permissions to ensure the security of the platform, while time and money are consumed within a reasonable range.

cs.CR↗

arXiv2020

Cross-lingual Machine Reading Comprehension with Language Branch Knowledge Distillation

Cross-lingual Machine Reading Comprehension (CLMRC) remains a challenging problem due to the lack of large-scale annotated datasets in low-source languages, such as Arabic, Hindi, and Vietnamese. Many previous approaches use translation data by translating from a rich-source language, such as English, to low-source languages as auxiliary supervision. However, how to effectively leverage translation data and reduce the impact of noise introduced by translation remains onerous. In this paper, we tackle this challenge and enhance the cross-lingual transferring performance by a novel augmentation approach named Language Branch Machine Reading Comprehension (LBMRC). A language branch is a group of passages in one single language paired with questions in all target languages. We train multiple machine reading comprehension (MRC) models proficient in individual language based on LBMRC. Then, we devise a multilingual distillation approach to amalgamate knowledge from multiple language branch models to a single model for all target languages. Combining the LBMRC and multilingual distillation can be more robust to the data noises, therefore, improving the model's cross-lingual ability. Meanwhile, the produced single multilingual model is applicable to all target languages, which saves the cost of training, inference, and maintenance for multiple models. Extensive experiments on two CLMRC benchmarks clearly show the effectiveness of our proposed method.

cs.CL↗

arXiv2020

BERT-EMD: Many-to-Many Layer Mapping for BERT Compression with Earth Mover's Distance

Pre-trained language models (e.g., BERT) have achieved significant success in various natural language processing (NLP) tasks. However, high storage and computational costs obstruct pre-trained language models to be effectively deployed on resource-constrained devices. In this paper, we propose a novel BERT distillation method based on many-to-many layer mapping, which allows each intermediate student layer to learn from any intermediate teacher layers. In this way, our model can learn from different teacher layers adaptively for various NLP tasks. %motivated by the intuition that different NLP tasks require different levels of linguistic knowledge contained in the intermediate layers of BERT. In addition, we leverage Earth Mover's Distance (EMD) to compute the minimum cumulative cost that must be paid to transform knowledge from teacher network to student network. EMD enables the effective matching for many-to-many layer mapping. %EMD can be applied to network layers with different sizes and effectively measures semantic distance between the teacher network and student network. Furthermore, we propose a cost attention mechanism to learn the layer weights used in EMD automatically, which is supposed to further improve the model's performance and accelerate convergence time. Extensive experiments on GLUE benchmark demonstrate that our model achieves competitive performance compared to strong competitors in terms of both accuracy and model compression.

cs.CL↗

arXiv2020

Improve Adversarial Robustness via Weight Penalization on Classification Layer

It is well-known that deep neural networks are vulnerable to adversarial attacks. Recent studies show that well-designed classification parts can lead to better robustness. However, there is still much space for improvement along this line. In this paper, we first prove that, from a geometric point of view, the robustness of a neural network is equivalent to some angular margin condition of the classifier weights. We then explain why ReLU type function is not a good choice for activation under this framework. These findings reveal the limitations of the existing approaches and lead us to develop a novel light-weight-penalized defensive method, which is simple and has a good scalability. Empirical results on multiple benchmark datasets demonstrate that our method can effectively improve the robustness of the network without requiring too much additional computation, while maintaining a high classification precision for clean data.

cs.LG↗

arXiv2020

Acoustic Metal

Metal reflects electromagnetic waves because of the large conductivity that is responsible for dissipation. During which the waves undergo a 180$^\circ$ phase change that is independent of the frequency. There is no counterpart material for acoustic waves. Here we show that by using an array of acoustic resonators with a designed high-density dissipative component, an "acoustic metal" can be realised that strongly couples with sound over a wide frequency range not otherwise attainable by conventional means. In particular, we show the acoustic Faraday cage effect that when used as a ring covering an air duct, 99% of the noise can be blocked without impeding the airflow. We further delineate the underlying volume requirement for an acoustic metal based on the constraint of the causality principle. Our findings complement the missing properties of acoustic materials and pave the way to the strong wave-material couplings that are critical for the applications as high-performance audio devices.

cond-mat.mtrl-sci↗

arXiv2020

Measurement of Conduction and Valence Bands g-factors in a Transition Metal Dichalcogenide Monolayer

The electron valley and spin degree of freedom in monolayer transition-metal dichalcogenides can be manipulated in optical and transport measurements performed in magnetic fields. The key parameter for determining the Zeeman splitting, namely the separate contribution of the electron and hole g-factor, is inaccessible in most measurements. Here we present an original method that gives access to the respective contribution of the conduction and valence band to the measured Zeeman splitting. It exploits the optical selection rules of exciton complexes, in particular the ones involving inter-valley phonons, avoiding strong renormalization effects that compromise single particle g-factor determination in transport experiments. These studies yield a direct determination of single band g factors. We measure gc1= 0.86, gc2=3.84 for the bottom (top) conduction bands and gv=6.1 for the valence band of monolayer WSe2. These measurements are helpful for quantitative interpretation of optical and transport measurements performed in magnetic fields. In addition the measured g-factors are valuable input parameters for optimizing band structure calculations of these 2D materials.

cond-mat.mtrl-sci↗

arXiv2020

Empirical Evaluation of Multi-task Learning in Deep Neural Networks for Natural Language Processing

Multi-Task Learning (MTL) aims at boosting the overall performance of each individual task by leveraging useful information contained in multiple related tasks. It has shown great success in natural language processing (NLP). Currently, a number of MLT architectures and learning mechanisms have been proposed for various NLP tasks. However, there is no systematic exploration and comparison of different MLT architectures and learning mechanisms for their strong performance in-depth. In this paper, we conduct a thorough examination of typical MTL methods on a broad range of representative NLP tasks. Our primary goal is to understand the merits and demerits of existing MTL methods in NLP tasks, thus devising new hybrid architectures intended to combine their strengths.

cs.CL↗

arXiv2020

Semi-Supervised Recognition under a Noisy and Fine-grained Dataset

Simi-Supervised Recognition Challenge-FGVC7 is a challenging fine-grained recognition competition. One of the difficulties of this competition is how to use unlabeled data. We adopted pseudo-tag data mining to increase the amount of training data. The other one is how to identify similar birds with a very small difference, especially those have a relatively tiny main-body in examples. We combined generic image recognition and fine-grained image recognition method to solve the problem. All generic image recognition models were training using PaddleClas . Using the combination of two different ways of deep recognition models, we finally won the third place in the competition.

cs.CV↗

arXiv2020

A Generic Network Compression Framework for Sequential Recommender Systems

Sequential recommender systems (SRS) have become the key technology in capturing user's dynamic interests and generating high-quality recommendations. Current state-of-the-art sequential recommender models are typically based on a sandwich-structured deep neural network, where one or more middle (hidden) layers are placed between the input embedding layer and output softmax layer. In general, these models require a large number of parameters (such as using a large embedding dimension or a deep network architecture) to obtain their optimal performance. Despite the effectiveness, at some point, further increasing model size may be harder for model deployment in resource-constraint devices, resulting in longer responding time and larger memory footprint. To resolve the issues, we propose a compressed sequential recommendation framework, termed as CpRec, where two generic model shrinking techniques are employed. Specifically, we first propose a block-wise adaptive decomposition to approximate the input and softmax matrices by exploiting the fact that items in SRS obey a long-tailed distribution. To reduce the parameters of the middle layers, we introduce three layer-wise parameter sharing schemes. We instantiate CpRec using deep convolutional neural network with dilated kernels given consideration to both recommendation accuracy and efficiency. By the extensive ablation studies, we demonstrate that the proposed CpRec can achieve up to 4$\sim$8 times compression rates in real-world SRS datasets. Meanwhile, CpRec is faster during training\inference, and in most cases outperforms its uncompressed counterpart.

cs.IR↗

arXiv2020

Amorphous Mo-Ta oxide nanotubes for long-term stable Mo oxide based supercapacitors

With a large-scale usage of portable electric appliances, a high demand for increasingly high density energy storage devices has emerged. MoO3 has, in principle, a large potential as negative electrode material in supercapacitive devices, due to high charge densities that can be obtained from its reversible redox reactions. Nevertheless, the extremely poor electrochemical stability of MoO3 in aqueous electrolytes prevents a practical use in high capacitance devices. In this work, we describe how to overcome this severe stability issue by forming amorphous molybdenum oxide/tantalum oxide nanotubes by anodic oxidation of a Mo-Ta alloy. The presence of a critical amount of Ta-oxide (> 20 at-%) prevents the electrochemical decay of the MoO3 phase and thus yields an extremely high stability. Due to the protection provided by tantalum oxide, no capacitance losses are measureable after 10000 charg-ing/discharging cycles.

physics.app-ph↗

arXiv2020

Convolutional Hierarchical Attention Network for Query-Focused Video Summarization

Previous approaches for video summarization mainly concentrate on finding the most diverse and representative visual contents as video summary without considering the user's preference. This paper addresses the task of query-focused video summarization, which takes user's query and a long video as inputs and aims to generate a query-focused video summary. In this paper, we consider the task as a problem of computing similarity between video shots and query. To this end, we propose a method, named Convolutional Hierarchical Attention Network (CHAN), which consists of two parts: feature encoding network and query-relevance computing module. In the encoding network, we employ a convolutional network with local self-attention mechanism and query-aware global attention mechanism to learns visual information of each shot. The encoded features will be sent to query-relevance computing module to generate queryfocused video summary. Extensive experiments on the benchmark dataset demonstrate the competitive performance and show the effectiveness of our approach.

cs.CV↗

arXiv2020

Discovering Protagonist of Sentiment with Aspect Reconstructed Capsule Network

Most recent existing aspect-term level sentiment analysis (ATSA) approaches combined various neural network models with delicately carved attention mechanisms built upon given aspect and context to generate refined sentence representations for better predictions. In these methods, aspect terms are always provided in both training and testing process which may degrade aspect-level analysis into sentence-level prediction. However, the annotated aspect term might be unavailable in real-world scenarios which may challenge the applicability of the existing methods. In this paper, we aim to improve ATSA by discovering the potential aspect terms of the predicted sentiment polarity when the aspect terms of a test sentence are unknown. We access this goal by proposing a capsule network based model named CAPSAR. In CAPSAR, sentiment categories are denoted by capsules and aspect term information is injected into sentiment capsules through a sentiment-aspect reconstruction procedure during the training. As a result, coherent patterns between aspects and sentimental expressions are encapsulated by these sentiment capsules. Experiments on three widely used benchmarks demonstrate these patterns have potential in exploring aspect terms from test sentence when only feeding the sentence to the model. Meanwhile, the proposed CAPSAR can clearly outperform SOTA methods in standard ATSA tasks.

cs.CL↗

arXiv2020

Valley Phonons and Exciton Complexes in a Monolayer Semiconductor

The coupling between spin, charge, and lattice degrees of freedom plays an important role in a wide range of fundamental phenomena. Monolayer semiconducting transitional metal dichalcogenides have emerged as an outstanding platform for studying these coupling effects because they possess unique spin-valley locking physics for hosting rich excitonic species and the reduced screening for strong Coulomb interactions. Here, we report the observation of multiple valley phonons, phonons with momentum vectors pointing to the corners of the hexagonal Brillouin zone, and the resulting exciton complexes in the monolayer semiconductor WSe2. From Lande g-factor and polarization analyses of photoluminescence peaks, we find that these valley phonons lead to efficient intervalley scattering of quasi particles in both exciton formation and relaxation. This leads to a series of photoluminescence peaks as valley phonon replicas of dark trions. Using identified valley phonons, we also uncovered an intervalley exciton near charge neutrality, and extract its short-range electron-hole exchange interaction to be about 10 meV. Our work not only identifies a number of previously unknown 2D excitonic species, but also shows that monolayer WSe2 is a prime candidate for studying interactions between spin, pseudospin, and zone-edge phonons.

cond-mat.mes-hall↗

arXiv2019

Improving Knowledge-aware Dialogue Generation via Knowledge Base Question Answering

Neural network models usually suffer from the challenge of incorporating commonsense knowledge into the open-domain dialogue systems. In this paper, we propose a novel knowledge-aware dialogue generation model (called TransDG), which transfers question representation and knowledge matching abilities from knowledge base question answering (KBQA) task to facilitate the utterance understanding and factual knowledge selection for dialogue generation. In addition, we propose a response guiding attention and a multi-step decoding strategy to steer our model to focus on relevant features for response generation. Experiments on two benchmark datasets demonstrate that our model has robust superiority over compared methods in generating informative and fluent dialogues. Our code is available at https://github.com/siat-nlp/TransDG.

cs.CL↗