arXiv subjects

Weidong Shi

Publications and source records attributed to Weidong Shi.

54 records · Page 3Linked to original sources

arXiv2021

Statistical Analysis of Perspective Scores on Hate Speech Detection

Hate speech detection has become a hot topic in recent years due to the exponential growth of offensive language in social media. It has proven that, state-of-the-art hate speech classifiers are efficient only when tested on the data with the same feature distribution as training data. As a consequence, model architecture plays the second role to improve the current results. In such a diverse data distribution relying on low level features is the main cause of deficiency due to natural bias in data. That's why we need to use high level features to avoid a biased judgement. In this paper, we statistically analyze the Perspective Scores and their impact on hate speech detection. We show that, different hate speech datasets are very similar when it comes to extract their Perspective Scores. Eventually, we prove that, over-sampling the Perspective Scores of a hate speech dataset can significantly improve the generalization performance when it comes to be tested on other hate speech datasets.

cs.CL↗

arXiv2021

Demystifying Pythia: A Survey of ChainLink Oracles Usage on Ethereum

Smart contracts are dependent on oracle systems for their adoption and usability. We perform an empirical study of oracle systems' usage trends and adoption metrics to provide better insight into the health of the smart contract ecosystem. We collect ChainLink usage data on the Ethereum network using a modified Ethereum client and running a full node. We analyze the collected data and present our findings and insights surrounding the usage trends, adoption metrics, oracle pricing and service quality associated with ChainLink on the Ethereum network.

cs.DC↗

arXiv2021

Nonlinear Blockchain Scalability: a Game-Theoretic Perspective

Recent advances in the blockchain research have been made in two important directions. One is refined resilience analysis utilizing game theory to study the consequences of selfish behaviors of users (miners), and the other is the extension from a linear (chain) structure to a non-linear (graphical) structure for performance improvements, such as IOTA and Graphcoin. The first question that comes to people's minds is what improvements that a blockchain system would see by leveraging these new advances. In this paper, we consider three major metrics for a blockchain system: full verification, scalability, and finality-duration. We { establish a formal framework and} prove that no blockchain system can achieve full verification, high scalability, and low finality-duration simultaneously. We observe that classical blockchain systems like Bitcoin achieves full verification and low finality-duration, Harmony and Ethereum 2.0 achieve low finality-duration and high scalability. As a complementary, we design a non-linear blockchain system that achieves full verification and scalability. We also establish, for the first time, the trade-off between scalability and finality-duration.

cs.CR↗

arXiv2021

EDSC: An Event-Driven Smart Contract Platform

This paper presents EDSC, a novel smart contract platform design based on the event-driven execution model as opposed to the traditionally employed transaction-driven execution model. We reason that such a design is a better fit for many emerging smart contract applications and is better positioned to address the scalability and performance challenges plaguing the smart contract ecosystem. We propose EDSC's design under the Ethereum framework, and the design can be easily adapted for other existing smart contract platforms. We have conducted implementation using Ethereum client and experiments where performance modeling results show on average 2.2 to 4.6 times reduced total latency of event triggered smart contracts, which demonstrates its effectiveness for supporting contracts that demand timely execution based on events. In addition, we discuss example use cases to demonstrate the design's utility and comment on its potential security dynamics.

cs.DC↗

arXiv2020

ProxylessKD: Direct Knowledge Distillation with Inherited Classifier for Face Recognition

Knowledge Distillation (KD) refers to transferring knowledge from a large model to a smaller one, which is widely used to enhance model performance in machine learning. It tries to align embedding spaces generated from the teacher and the student model (i.e. to make images corresponding to the same semantics share the same embedding across different models). In this work, we focus on its application in face recognition. We observe that existing knowledge distillation models optimize the proxy tasks that force the student to mimic the teacher's behavior, instead of directly optimizing the face recognition accuracy. Consequently, the obtained student models are not guaranteed to be optimal on the target task or able to benefit from advanced constraints, such as large margin constraints (e.g. margin-based softmax). We then propose a novel method named ProxylessKD that directly optimizes face recognition accuracy by inheriting the teacher's classifier as the student's classifier to guide the student to learn discriminative embeddings in the teacher's embedding space. The proposed ProxylessKD is very easy to implement and sufficiently generic to be extended to other tasks beyond face recognition. We conduct extensive experiments on standard face recognition benchmarks, and the results demonstrate that ProxylessKD achieves superior performance over existing knowledge distillation methods.

cs.CV↗

arXiv2020

Vulnerability of Face Recognition Systems Against Composite Face Reconstruction Attack

Rounding confidence score is considered trivial but a simple and effective countermeasure to stop gradient descent based image reconstruction attacks. However, its capability in the face of more sophisticated reconstruction attacks is an uninvestigated research area. In this paper, we prove that, the face reconstruction attacks based on composite faces can reveal the inefficiency of rounding policy as countermeasure. We assume that, the attacker takes advantage of face composite parts which helps the attacker to get access to the most important features of the face or decompose it to the independent segments. Afterwards, decomposed segments are exploited as search parameters to create a search path to reconstruct optimal face. Face composition parts enable the attacker to violate the privacy of face recognition models even with a blind search. However, we assume that, the attacker may take advantage of random search to reconstruct the target face faster. The algorithm is started with random composition of face parts as initial face and confidence score is considered as fitness value. Our experiments show that, since the rounding policy as countermeasure can't stop the random search process, current face recognition systems are extremely vulnerable against such sophisticated attacks. To address this problem, we successfully test Face Detection Score Filtering (FDSF) as a countermeasure to protect the privacy of training data against proposed attack.

cs.CV↗

arXiv2020

Towards Stable Imbalanced Data Classification via Virtual Big Data Projection

Virtual Big Data (VBD) proved to be effective to alleviate mode collapse and vanishing generator gradient as two major problems of Generative Adversarial Neural Networks (GANs) very recently. In this paper, we investigate the capability of VBD to address two other major challenges in Machine Learning including deep autoencoder training and imbalanced data classification. First, we prove that, VBD can significantly decrease the validation loss of autoencoders via providing them a huge diversified training data which is the key to reach better generalization to minimize the over-fitting problem. Second, we use the VBD to propose the first projection-based method called cross-concatenation to balance the skewed class distributions without over-sampling. We prove that, cross-concatenation can solve uncertainty problem of data driven methods for imbalanced classification.

cs.LG↗

arXiv2020

Computational Complexity Characterization of Protecting Elections from Bribery

The bribery problem in election has received considerable attention in the literature, upon which various algorithmic and complexity results have been obtained. It is thus natural to ask whether we can protect an election from potential bribery. We assume that the protector can protect a voter with some cost (e.g., by isolating the voter from potential bribers). A protected voter cannot be bribed. Under this setting, we consider the following bi-level decision problem: Is it possible for the protector to protect a proper subset of voters such that no briber with a fixed budget on bribery can alter the election result? The goal of this paper is to give a full picture on the complexity of protection problems. We give an extensive study on the protection problem and provide algorithmic and complexity results. Comparing our results with that on the bribery problems, we observe that the protection problem is in general significantly harder. Indeed, it becomes $\sum_{p}^2$-complete even for very restricted special cases, while most bribery problems lie in NP. However, it is not necessarily the case that the protection problem is always harder. Some of the protection problems can still be solved in polynomial time, while some of them remain as hard as the bribery problem under the same setting.

cs.CC↗

arXiv2020

One-Shot GAN Generated Fake Face Detection

Fake face detection is a significant challenge for intelligent systems as generative models become more powerful every single day. As the quality of fake faces increases, the trained models become more and more inefficient to detect the novel fake faces, since the corresponding training data is considered outdated. In this case, robust One-Shot learning methods is more compatible with the requirements of changeable training data. In this paper, we propose a universal One-Shot GAN generated fake face detection method which can be used in significantly different areas of anomaly detection. The proposed method is based on extracting out-of-context objects from faces via scene understanding models. To do so, we use state of the art scene understanding and object detection methods as a pre-processing tool to detect the weird objects in the face. Second, we create a bag of words given all the detected out-of-context objects per all training data. This way, we transform each image into a sparse vector where each feature represents the confidence score related to each detected object in the image. Our experiments show that, we can discriminate fake faces from real ones in terms of out-of-context features. It means that, different sets of objects are detected in fake faces comparing to real ones when we analyze them with scene understanding and object detection models. We prove that, the proposed method can outperform previous methods based on our experiments on Style-GAN generated fake faces.

cs.CV↗

arXiv2020

Deep Synthetic Minority Over-Sampling Technique

Synthetic Minority Over-sampling Technique (SMOTE) is the most popular over-sampling method. However, its random nature makes the synthesized data and even imbalanced classification results unstable. It means that in case of running SMOTE n different times, n different synthesized in-stances are obtained with n different classification results. To address this problem, we adapt the SMOTE idea in deep learning architecture. In this method, a deep neural network regression model is used to train the inputs and outputs of traditional SMOTE. Inputs of the proposed deep regression model are two randomly chosen data points which are concatenated to form a double size vector. The outputs of this model are corresponding randomly interpolated data points between two randomly chosen vectors with original dimension. The experimental results show that, Deep SMOTE can outperform traditional SMOTE in terms of precision, F1 score and Area Under Curve (AUC) in majority of test cases.

cs.LG↗

arXiv2020

Hybrid Cryptocurrency Pump and Dump Detection

Increasingly growing Cryptocurrency markets have become a hive for scammers to run pump and dump schemes which is considered as an anomalous activity in exchange markets. Anomaly detection in time series is challenging since existing methods are not sufficient to detect the anomalies in all contexts. In this paper, we propose a novel hybrid pump and dump detection method based on distance and density metrics. First, we propose a novel automatic thresh-old setting method for distance-based anomaly detection. Second, we propose a novel metric called density score for density-based anomaly detection. Finally, we exploit the combination of density and distance metrics successfully as a hybrid approach. Our experiments show that, the proposed hybrid approach is reliable to detect the majority of alleged P & D activities in top ranked exchange pairs by outperforming both density-based and distance-based methods.

cs.AI↗

arXiv2020

Algebraic multigrid block preconditioning for multi-group radiation diffusion equations

The paper focuses on developing and studying efficient block preconditioners based on classical algebraic multigrid for the large-scale sparse linear systems arising from the fully coupled and implicitly cell-centered finite volume discretization of multi-group radiation diffusion equations, whose coefficient matrices can be rearranged into the $(G+2)\times(G+2)$ block form, where $G$ is the number of energy groups. The preconditioning techniques are based on the monolithic classical algebraic multigrid method, physical-variable based coarsening two-level algorithm and two types of block Schur complement preconditioners. The classical algebraic multigrid is applied to solve the subsystems that arise in the last three block preconditioners. The coupling strength and diagonal dominance are further explored to improve performance. We use representative one-group and twenty-group linear systems from capsule implosion simulations to test the robustness, efficiency, strong and weak parallel scaling properties of the proposed methods. Numerical results demonstrate that block preconditioners lead to mesh- and problem-independent convergence, and scale well both algorithmically and in parallel.

math.NA↗

arXiv2019

New Bounds on Augmenting Steps of Block-structured Integer Programs

We consider 4-block $n$-fold integer programs, whose constraint matrix consists of $n$ copies of small matrices $A$, $B$, and $D$, and one copy of $C$, in a specific block structure. All existing algorithms along this line of research follows an iterative augmentation framework, which relies on the so-called Graver basis of the constraint matrix that constitutes a set of fundamental augmenting steps. Bounding the $\ell_1$- or $\ell_\infty$-norm of elements of the Graver basis is the key to these algorithms. Hemmecke et al.~[Math. Prog. 2014] showed that 4-block $n$-fold IP has Graver elements of $\ell_\infty$-norm at most $O_{FPT}(n^{2^{s_{D}}})$, leading to an algorithm with a similar runtime; here, $s_{D}$ is the number of rows of matrix $D$ and $ O_{FPT}(1)$ hides a multiplicative factor that is only dependent on the small matrices $A,B,C,D$. We prove that the $\ell_{\infty}$-norm of the Graver elements of 4-block $n$-fold IP is upper bounded by $O_{FPT}(n^{s_{D}})$, improving significantly over the previous bound $O_{FPT} (n^{2^{s_{D}}})$. We also provide a matching lower bound of $Ω(n^{s_{D}})$ which even holds for arbitrary non-zero lattice elements, ruling out augmenting algorithm relying on even more restricted notions of augmentation than the Graver basis. We then consider a special case of 4-block $n$-fold in which $C$ is a zero matrix, called 3-block $n$-fold IP. We show that while even there the $\ell_{\infty}$-norm of its Graver elements is $Ω(n^{s_{D}})$, there exists a different decomposition into lattice elements whose $\ell_{\infty}$-norm is bounded by $ O_{FPT}(1)$, which allows us to provide improved upper bounds on the $\ell_{\infty}$-norm of Graver elements for 3-block $n$-fold IP.

cs.DS↗

arXiv2019

Private and Atomic Exchange of Assets over Zero Knowledge Based Payment Ledger

Bitcoin brings a new type of digital currency that does not rely on a central system to maintain transactions. By benefiting from the concept of decentralized ledger, users who do not know or trust each other can still conduct transactions in a peer-to-peer manner. Inspired by Bitcoin, other cryptocurrencies were invented in recent years such as Ethereum, Dash, Zcash, Monero, Grin, etc. Some of these focus on enhancing privacy for instance crypto note or systems that apply the similar concept of encrypted notes used for transactions to enhance privacy (e.g., Zcash, Monero). However, there are few mechanisms to support the exchange of privacy-enhanced notes or assets on the chain, and at the same time preserving the privacy of the exchange operations. Existing approaches for fair exchanges of assets with privacy mostly rely on off-chain/side-chain, escrow or centralized services. Thus, we propose a solution that supports oblivious and privacy-protected fair exchange of crypto notes or privacy enhanced crypto assets. The technology is demonstrated by extending zero-knowledge based crypto notes. To address "privacy" and "multi-currency", we build a new zero-knowledge proving system and extend note format with new property to represent various types of tokenized assets or cryptocurrencies. By extending the payment protocol, exchange operations are realized through privacy enhanced transactions (e.g., shielded transactions). Based on the possible scenarios during the exchange operation, we add new constraints and conditions to the zero-knowledge proving system used for validating transactions publicly.

cs.CR↗

arXiv2018

Efficient Public Blockchain Client for Lightweight Users

Public blockchains provide a decentralized method for storing transaction data and have many applications in different sectors. In order for users to track transactions, a simple method is to let them keep a local copy of the entire public ledger. Since the size of the ledger keeps growing, this method becomes increasingly less practical, especially for lightweight users such as IoT devices and smartphones. In order to cope with the problem, several solutions have been proposed to reduce the storage burden. However, existing solutions either achieve a limited storage reduction (e.g., simple payment verification), or rely on some strong security assumption (e.g., the use of trusted server). In this paper, we propose a new approach to solving the problem. Specifically, we propose an \underline{e}fficient verification protocol for \underline{p}ublic \underline{b}lock\underline{c}hains, or EPBC for short. EPBC is particularly suitable for lightweight users, who only need to store a small amount of data that is {\it independent of} the size of the blockchain. We analyze EPBC's performance and security, and discuss its integration with existing public ledger systems. Experimental results confirm that EPBC is practical for lightweight users.

cs.CR↗

arXiv2018

Election with Bribed Voter Uncertainty: Hardness and Approximation Algorithm

Bribery in election (or computational social choice in general) is an important problem that has received a considerable amount of attention. In the classic bribery problem, the briber (or attacker) bribes some voters in attempting to make the briber's designated candidate win an election. In this paper, we introduce a novel variant of the bribery problem, "Election with Bribed Voter Uncertainty" or BVU for short, accommodating the uncertainty that the vote of a bribed voter may or may not be counted. This uncertainty occurs either because a bribed voter may not cast its vote in fear of being caught, or because a bribed voter is indeed caught and therefore its vote is discarded. As a first step towards ultimately understanding and addressing this important problem, we show that it does not admit any multiplicative $O(1)$-approximation algorithm modulo standard complexity assumptions. We further show that there is an approximation algorithm that returns a solution with an additive-$ε$ error in FPT time for any fixed $ε$.

cs.DS↗

arXiv2018

Toward Efficient Breast Cancer Diagnosis and Survival Prediction Using L-Perceptron

Breast cancer is the most frequently reported cancer type among the women around the globe and beyond that it has the second highest female fatality rate among all cancer types. Despite all the progresses made in prevention and early intervention, early prognosis and survival prediction rates are still unsatisfactory. In this paper, we propose a novel type of perceptron called L-Perceptron which outperforms all the previous supervised learning methods by reaching 97.42 \% and 98.73 \% in terms of accuracy and sensitivity, respectively in Wisconsin Breast Cancer dataset. Experimental results on Haberman's Breast Cancer Survival dataset, show the superiority of proposed method by reaching 75.18 \% and 83.86 \% in terms of accuracy and F1 score, respectively. The results are the best reported ones obtained in 10-fold cross validation in absence of any preprocessing or feature selection.

cs.LG↗

arXiv2018

Next Generation of Star Patterns

In this paper we present two new ideas for generating star patterns and filling the gaps during the tile operation. Firstly, we introduce a novel parametric method based on concentric circles for generating stars and rosettes. Using proposed method, completely different stars and rosettes and a set of new and complex star patterns convert to each other only by changing nine parameters. Secondly, we demonstrate how three equal tangent circles can be used as a base for generating tile elements. For this reason a surrounded circle is created among tangent circles, which represents the gaps in hexagonal packing. Afterwards, we use our first idea for filling the tangent circles and surrounded circle. This parametric approach can be used for generating infinite new star patterns, which some of them will be presented in result section.Two Android apps of proposed method called Starking and Tilerking are available in Google app store.

cs.GR↗