arXiv subjects

Wei Lu

Publications and source records attributed to Wei Lu.

At least 307 records · Page 17Linked to original sources

arXiv2021

Detection of Deepfake Videos Using Long Distance Attention

With the rapid progress of deepfake techniques in recent years, facial video forgery can generate highly deceptive video contents and bring severe security threats. And detection of such forgery videos is much more urgent and challenging. Most existing detection methods treat the problem as a vanilla binary classification problem. In this paper, the problem is treated as a special fine-grained classification problem since the differences between fake and real faces are very subtle. It is observed that most existing face forgery methods left some common artifacts in the spatial domain and time domain, including generative defects in the spatial domain and inter-frame inconsistencies in the time domain. And a spatial-temporal model is proposed which has two components for capturing spatial and temporal forgery traces in global perspective respectively. The two components are designed using a novel long distance attention mechanism. The one component of the spatial domain is used to capture artifacts in a single frame, and the other component of the time domain is used to capture artifacts in consecutive frames. They generate attention maps in the form of patches. The attention method has a broader vision which contributes to better assembling global information and extracting local statistic information. Finally, the attention maps are used to guide the network to focus on pivotal parts of the face, just like other fine-grained classification methods. The experimental results on different public datasets demonstrate that the proposed method achieves the state-of-the-art performance, and the proposed long distance attention method can effectively capture pivotal parts for face forgery.

cs.CV↗

arXiv2021

Generative Adversarial Neural Architecture Search

Despite the empirical success of neural architecture search (NAS) in deep learning applications, the optimality, reproducibility and cost of NAS schemes remain hard to assess. In this paper, we propose Generative Adversarial NAS (GA-NAS) with theoretically provable convergence guarantees, promoting stability and reproducibility in neural architecture search. Inspired by importance sampling, GA-NAS iteratively fits a generator to previously discovered top architectures, thus increasingly focusing on important parts of a large search space. Furthermore, we propose an efficient adversarial learning approach, where the generator is trained by reinforcement learning based on rewards provided by a discriminator, thus being able to explore the search space without evaluating a large number of architectures. Extensive experiments show that GA-NAS beats the best published results under several cases on three public NAS benchmarks. In the meantime, GA-NAS can handle ad-hoc search constraints and search spaces. We show that GA-NAS can be used to improve already optimized baselines found by other NAS methods, including EfficientNet and ProxylessNAS, in terms of ImageNet accuracy or the number of parameters, in their original search space.

cs.LG↗

arXiv2021

Revenue Maximization in Incentivized Social Advertising

Incentivized social advertising, an emerging marketing model, provides monetization opportunities not only to the owners of the social networking platforms but also to their influential users by offering a "cut" on the advertising revenue. We consider a social network (the host) that sells ad-engagements to advertisers by inserting their ads, in the form of promoted posts, into the feeds of carefully selected "initial endorsers" or seed users: these users receive monetary incentives in exchange for their endorsements. The endorsements help propagate the ads to the feeds of their followers. In this context, the problem for the host is is to allocate ads to influential users, taking into account the propensity of ads for viral propagation, and carefully apportioning the monetary budget of each of the advertisers between incentives to influential users and ad-engagement costs, with the rational goal of maximizing its own revenue. We consider a monetary incentive for the influential users, which is proportional to their influence potential. We show that revenue maximization in incentivized social advertising corresponds to the problem of monotone submodular function maximization, subject to a partition matroid constraint on the ads-to-seeds allocation, and submodular knapsack constraints on the advertisers' budgets. This problem is NP-hard and we devise 2 greedy algorithms with provable approximation guarantees, which differ in their sensitivity to seed user incentive costs. Our approximation algorithms require repeatedly estimating the expected marginal gain in revenue as well as in advertiser payment. By exploiting a connection to the recent advances made in scalable estimation of expected influence spread, we devise efficient and scalable versions of the greedy algorithms.

cs.SI↗

arXiv2021

High efficiency uniform wakefield acceleration of a positron beam using stable asymmetric mode in a hollow channel plasma

Plasma wakefield acceleration in the blowout regime is particularly promising for high-energy acceleration of electron beams because of its potential to simultaneously provide large acceleration gradients and high energy transfer efficiency while maintaining excellent beam quality. However, no equivalent regime for positron acceleration in plasma wakes has been discovered to-date. We show that after a short propagation distance, an asymmetric electron beam drives a stable wakefield in a hollow plasma channel that can be both accelerating and focusing for a positron beam. A high charge positron bunch placed at a suitable distance behind the drive bunch can beam-load or flatten the longitudinal wakefield and enhance the transverse focusing force, leading to high-efficiency and narrow energy spread acceleration of the positrons. Three-dimensional quasi-static particle-in-cell (PIC) simulations show that over 30% energy extraction efficiency from the wake to the positrons and 1% level energy spread can be simultaneously obtained, and further optimization is feasible.

physics.plasm-ph↗

arXiv2021

Ultralow Threshold, Single-Mode InGaAs/GaAs Multi-Quantum Disk Nanowire Lasers

We present single-mode nanowire (NW) lasers with ultralow threshold in the near-infrared spectral range. To ensure the single-mode operation, the NW diameter and length are reduced specifically to minimize the longitudinal and transverse modes of the NW cavity. Increased optical losses and reduced gain volume by the dimension reduction are compensated by excellent NW morphology and InGaAs/GaAs multi-quantum disks. At 5 K, a threshold low as 1.6 μJ/cm2 per pulse is achieved with a resulting quality factor exceeding 6400. By further passivating the NW with an AlGaAs shell to suppress surface non-radiative recombination, single-mode lasing operation is obtained with a threshold of only 48 μJ/cm2 per pulse at room temperature with a high characteristic temperature of 223 K and power output of ~ 0.9 μW. These single-mode, ultralow threshold, high power output NW lasers are promising for the development of near-infrared nanoscale coherent light sources for integrated photonic circuits, sensing, and spectroscopy.

physics.optics↗

arXiv2021

A parallel-network continuous quantitative trading model with GARCH and PPO

It is a difficult task for both professional investors and individual traders continuously making profit in stock market. With the development of computer science and deep reinforcement learning, Buy\&Hold (B\&H) has been oversteped by many artificial intelligence trading algorithms. However, the information and process are not enough, which limit the performance of reinforcement learning algorithms. Thus, we propose a parallel-network continuous quantitative trading model with GARCH and PPO to enrich the basical deep reinforcement learning model, where the deep learning parallel network layers deal with 3 different frequencies data (including GARCH information) and proximal policy optimization (PPO) algorithm interacts actions and rewards with stock trading environment. Experiments in 5 stocks from Chinese stock market show our method achieves more extra profit comparing with basical reinforcement learning methods and bench models.

q-fin.TR↗

arXiv2021

Aspect Based Sentiment Analysis with Aspect-Specific Opinion Spans

Aspect based sentiment analysis, predicting sentiment polarity of given aspects, has drawn extensive attention. Previous attention-based models emphasize using aspect semantics to help extract opinion features for classification. However, these works are either not able to capture opinion spans as a whole, or not able to capture variable-length opinion spans. In this paper, we present a neat and effective structured attention model by aggregating multiple linear-chain CRFs. Such a design allows the model to extract aspect-specific opinion spans and then evaluate sentiment polarity by exploiting the extracted opinion features. The experimental results on four datasets demonstrate the effectiveness of the proposed model, and our analysis demonstrates that our model can capture aspect-specific opinion spans.

cs.CL↗

arXiv2021

Better Feature Integration for Named Entity Recognition

It has been shown that named entity recognition (NER) could benefit from incorporating the long-distance structured information captured by dependency trees. We believe this is because both types of features - the contextual information captured by the linear sequences and the structured information captured by the dependency trees may complement each other. However, existing approaches largely focused on stacking the LSTM and graph neural networks such as graph convolutional networks (GCNs) for building improved NER models, where the exact interaction mechanism between the two types of features is not very clear, and the performance gain does not appear to be significant. In this work, we propose a simple and robust solution to incorporate both types of features with our Synergized-LSTM (Syn-LSTM), which clearly captures how the two types of features interact. We conduct extensive experiments on several standard datasets across four languages. The results demonstrate that the proposed model achieves better performance than previous approaches while requiring fewer parameters. Our further analysis demonstrates that our model can capture longer dependencies compared with strong baselines.

cs.CL↗

arXiv2021

A unified approach for covariance matrix estimation under Stein loss

In this paper, we address the problem of estimating a covariance matrix of a multivariate Gaussian distribution, relative to a Stein loss function, from a decision theoretic point of view. We investigate the case where the covariance matrix is invertible and the case when it is non--invertible in a unified approach.

math.ST↗

arXiv2021

A new method for constructing linear codes with small hulls

The hull of a linear code over finite fields is the intersection of the code and its dual, which was introduced by Assmus and Key. In this paper, we develop a method to construct linear codes with trivial hull ( LCD codes) and one-dimensional hull by employing the positive characteristic analogues of Gauss sums. These codes are quasi-abelian, and sometimes doubly circulant. Some sufficient conditions for a linear code to be an LCD code (resp. a linear code with one-dimensional hull) are presented. It is worth mentioning that we present a lower bound on the minimum distances of the constructed linear codes. As an application, using these conditions, we obtain some optimal or almost optimal LCD codes (resp. linear codes with one-dimensional hull) with respect to the online Database of Grassl.

cs.IT↗

arXiv2021

Position-Aware Tagging for Aspect Sentiment Triplet Extraction

Aspect Sentiment Triplet Extraction (ASTE) is the task of extracting the triplets of target entities, their associated sentiment, and opinion spans explaining the reason for the sentiment. Existing research efforts mostly solve this problem using pipeline approaches, which break the triplet extraction process into several stages. Our observation is that the three elements within a triplet are highly related to each other, and this motivates us to build a joint model to extract such triplets using a sequence tagging approach. However, how to effectively design a tagging approach to extract the triplets that can capture the rich interactions among the elements is a challenging research question. In this work, we propose the first end-to-end model with a novel position-aware tagging scheme that is capable of jointly extracting the triplets. Our experimental results on several existing datasets show that jointly capturing elements in the triplet using our approach leads to improved performance over the existing approaches. We also conducted extensive experiments to investigate the model effectiveness and robustness.

cs.CL↗

arXiv2021

On 2-digit and 3-digit Kaprekar's Routine

Kaprekar's Routine is an iteration process that, with each iteration, sorts the digits of a number in descending and ascending orders and uses their difference for the next iteration. In this paper, we successfully address the structure of the fixed periodic sets and the maximum step of iteration to enter a fixed set in the three-digit case and two-digit case, with a focus on the latter. With three-digit, there exists a unique fixed set of the cardinality of 1 or 2, depending on the parity of m, and the maximum step is determined by m. With two-digit, we offer some powerful properties of the fixed sets and a mechanical method to calculate the fixed sets. We also prove that the number of 2-factors in m+1 determines the maximum step (but has completely no influence on the structure of the fixed sets). This continues Atsushi Yamagami's previous studies in 2018 (published in Journal of Number Theory) and completes it with a general solution.

math.NT↗

arXiv2021

Simulation of an imaging system for internal contamination of lungs using MPA-MURA coded aperture collimator

The nuclides inhaled during nuclear accidents usually cause internal contamination of the lungs with low activity. Although a parallel-hole imaging system, which is widely used in medical gamma cameras, has a high resolution and good image quality, owing to its extremely low detection efficiency, it remains difficult to obtain images of inhaled lung contamination. In this study, the Monte Carlo method was used to study the internal lung contamination imaging using the MPA-MURA coded-aperture collimator. The imaging system consisted of an adult male lung model, with a mosaicked, pattern-centered, and anti-symmetric MURA coded-aperture collimator model and a CsI(Tl) detector model. The MLEM decoding algorithm was used to reconstruct the internal contamination image, and the complementary imaging method was used to reduce the number of artifacts. The full width at half maximum of the I-131 point source image reconstructed by the mosaicked, pattern-centered, and anti-symmetric Modified uniformly redundant array (MPA-MURA) coded-aperture imaging reached 2.51 mm, and the signal-to-noise ratio of the simplified respiratory tract source (I-131) image reconstructed through MPA-MURA coded-aperture imaging was 3.98 dB. Although the spatial resolution of MPA-MURA coded aperture imaging is not as good as that of parallel-hole imaging, the detection efficiency of PMA-MURA coded-aperture imaging is two orders of magnitude higher than that of parallel hole collimator imaging. Considering the low activity level of internal lung contamination caused by nuclear accidents, PMA-MURA coded-aperture imaging has significant potential for the development of lung contamination imaging.

physics.med-ph↗

arXiv2020

Time to Transfer: Predicting and Evaluating Machine-Human Chatting Handoff

Is chatbot able to completely replace the human agent? The short answer could be - "it depends...". For some challenging cases, e.g., dialogue's topical spectrum spreads beyond the training corpus coverage, the chatbot may malfunction and return unsatisfied utterances. This problem can be addressed by introducing the Machine-Human Chatting Handoff (MHCH), which enables human-algorithm collaboration. To detect the normal/transferable utterances, we propose a Difficulty-Assisted Matching Inference (DAMI) network, utilizing difficulty-assisted encoding to enhance the representations of utterances. Moreover, a matching inference mechanism is introduced to capture the contextual matching features. A new evaluation metric, Golden Transfer within Tolerance (GT-T), is proposed to assess the performance by considering the tolerance property of the MHCH. To provide insights into the task and validate the proposed model, we collect two new datasets. Extensive experimental results are presented and contrasted against a series of baseline models to demonstrate the efficacy of our model on MHCH.

cs.CL↗

arXiv2020

Minimal Linear Codes Constructed from Functions

In this paper, we consider minimal linear codes in a general construction of linear codes from q-ary functions. First, we give the sufficient and necessary condition for codewords to be minimal. Second, as an application, we present four constructions of minimal linear codes which contained some recent results as special cases.

cs.IT↗

arXiv2020

Ultra-short pulse generation from mid-IR to THz range using plasma wakes and relativistic ionization fronts

This paper discusses numerical and experimental results on frequency downshifting and upshifting of a 10 $μ$m infrared laser to cover the entire wavelength (frequency) range from $λ$=1-150 $μ$m ($ν$=300-2 THz) using two different plasma techniques. The first plasma technique utilizes frequency downshifting of the drive laser pulse in a nonlinear plasma wake. Based on this technique, we have proposed and demonstrated that in a tailored plasma structure multi-millijoule energy, single-cycle, long-wavelength IR (3-20 $μ$m) pulses can be generated by using an 810 nm Ti:sapphire drive laser. Here we extend this idea to the THz frequency regime. We show that sub-joule, terawatts, single-cycle terahertz (2-12 THz, or 150-25 $μ$m) pulses can be generated by replacing the drive laser with a picosecond 10 $μ$m CO$_2$ laser and a different shaped plasma structure. The second plasma technique employs frequency upshifting by colliding a CO$_2$ laser with a rather sharp relativistic ionization front created by ionization of a gas in less than half cycle (17 fs) of the CO$_2$ laser. Even though the electrons in the ionization front carry no energy, the frequency of the CO$_2$ laser can be upshifted due to the relativistic Doppler effect as the CO$_2$ laser pulse enters the front. The wavelength can be tuned from 1-10 $μ$m by simply changing the electron density of the front. While the upshifted light with $5 <λ(μ$m$)< 10$ propagates in the forward direction, that with $1 <λ(μ$m$)< 5$ is back-reflected. These two plasma techniques seem extremely promising for covering the entire molecular fingerprint region.

physics.plasm-ph↗

arXiv2020

Measurements of the growth and saturation of electron Weibel instability in optical-field ionized plasmas

The temporal evolution of the magnetic field associated with electron thermal Weibel instability in optical-field ionized plasmas is measured using ultrashort (1.8 ps), relativistic (45 MeV) electron bunches from a linear accelerator. The self-generated magnetic fields are found to self-organize into a quasi-static structure consistent with a helicoid topology within a few ps and such a structure lasts for tens of ps in underdense plasmas. The measured growth rate agrees well with that predicted by the kinetic theory of plasmas taking into account collisions. Magnetic trapping is identified as the dominant saturation mechanism.

physics.plasm-ph↗

arXiv2020

Exploring Direct Citations between Citing Publications

This paper defines and explores the direct citations between citing publications (DCCPs) of a publication. We construct an ego-centered citation network for each paper that contains all of its citing papers and itself, as well as the citation relationships among them. By utilizing a large-scale scholarly dataset from the computer science field in the Microsoft Academic Graph (MAG-CS) dataset, we find that DCCPs exist universally in medium and highly cited papers. For those papers that have DCCPs, DCCPs do occur frequently; highly cited papers tend to contain more DCCPs than others. Meanwhile, the number of DCCPs of papers published in different years does not vary dramatically. The current paper also discusses the relationship between DCCPs and some indirect citation relationships (e.g., co-citation and bibliographic coupling).

cs.DL↗