Search

Among the most famous algorithms for solving classification problems are support vector machines (SVMs), which find a separating hyperplane for a set of labeled data points. In some applications, however, labels are only available for a subset of points. Furthermore, this subset can be non-representative, e.g., due to self-selection in a survey. Semi-supervised SVMs tackle the setting of labeled and unlabeled data and can often improve the reliability of the results. Moreover, additional information about the size of the classes can be available from undisclosed sources. We propose a mixed-integer quadratic optimization (MIQP) model that covers the setting of labeled and unlabeled data points as well as the overall number of points in each class. Since the MIQP's solution time rapidly grows as the number of variables increases, we introduce an iterative clustering approach to reduce the model's size. Moreover, we present an update rule for the required big-M values, prove the correctness of the iterative clustering method as well as derive tailored dimension-reduction and warm-starting techniques. Our numerical results show that our approach leads to a similar accuracy and precision than the MIQP formulation but at much lower computational cost. Thus, we can solve larger problems. With respect to the original SVM formulation, we observe that our approach has even better accuracy and precision for biased samples. [ABSTRACT FROM AUTHOR]

Unsupervised learning methods are well established in the area of anomaly detection and achieve state of the art performances on outlier datasets. Outliers play a significant role, since they bear the potential to distort the predictions of a machine learning algorithm on a given dataset. Especially among PCA-based methods, outliers have an additional destructive potential regarding the result: they may not only distort the orientation and translation of the principal components, they also make it more complicated to detect outliers. To address this problem, we propose the robust outlier detection algorithm CoMadOut, which satisfies two required properties: (1) being robust towards outliers and (2) detecting them. Our CoMadOut outlier detection variants using comedian PCA define, dependent on its variant, an inlier region with a robust noise margin by measures of in-distribution (variant CMO) and optimized scores by measures of out-of-distribution (variants CMO*), e.g. kurtosis-weighting by CMO+k. These measures allow distribution based outlier scoring for each principal component, and thus, an appropriate alignment of the degree of outlierness between normal and abnormal instances. Experiments comparing CoMadOut with traditional, deep and other comparable robust outlier detection methods showed that the performance of the introduced CoMadOut approach is competitive to well established methods related to average precision (AP), area under the precision recall curve (AUPRC) and area under the receiver operating characteristic (AUROC) curve. In summary our approach can be seen as a robust alternative for outlier detection tasks. [ABSTRACT FROM AUTHOR]

Given any deep fully connected neural network, initialized with random Gaussian parameters, we bound from above the quadratic Wasserstein distance between its output distribution and a suitable Gaussian process. Our explicit inequalities indicate how the hidden and output layers sizes affect the Gaussian behaviour of the network and quantitatively recover the distributional convergence results in the wide limit, i.e., if all the hidden layers sizes become large. [ABSTRACT FROM AUTHOR]

The issue of traffic mismanagement is getting bad to worse as the trend of urbanization is increasing and posing significant threats to the urban areas. This paper addresses this issue of traffic mismanagement by proposing a Fog-Cloud centric Internet of Things (IoT)-based collaborative framework that enables machine learning-based situation-aware traffic management. This framework monitors the traffic dynamics at the fog layer for providing real-time analytics to the cloud and enabling the traffic situation-aware routing of the vehicles. Whereas the framework at the cloud layer determines the situation-aware analytics by predicting Waiting Time (WaT), Waiting Queue Length (WQueL) and next green phase duration (G) for each Traffic Movement Signal Point (TMSP) based on the real-time data and analytics provided by the fog layer. The road-infrastructure hosted fog layer classifies the In-schedule Reachability Status (IRS) that determines the ability of the vehicle to reach the respective TMSP down the lane before the start of the next green phase. After classifying all such vehicles on a lane by the fog layer, the cloud layer predicts the optimal duration of the next green phase at that respective TMSP to reduce the traffic congestion on the respective junction point and subsequently provides these area-wide analytics to the vehicle hosted fog nodes and traffic controllers on the respective junction points. The vehicles use these analytics for enabling situation-aware vehicle routing functionality by selecting a time optimized path and enabling balanced traffic load on all possible paths. The framework employs Logistic Regression (LR) for IRS classification at the fog layer and Artificial Neural Network (ANN) for traffic prediction at the respective TMSP at the cloud layer. The result depiction acknowledges LR's efficiency compared to other employed classifiers in terms of various statistical parameters. The optimal performance of ANN at the learning rate (LeR) of 0.1, momentum rate (MoR) of 0.95, and 500 epochs depict the prediction efficiency of ANN as compared to other used prediction approach. The contribution of this paper is two functionalities-based working of fog computing for providing real-time and situation-aware traffic analytics, adaptive traffic movement phase planning, time optimized navigation, and optimal traffic load balancing. [ABSTRACT FROM AUTHOR]

We develop a data-driven machine learning (ML) turbulence model and conduct a comparative study with representative approaches employed in the literature. The new approach is based on a recently proposed method to extract, from the mean velocity field high-fidelity data, the essential information to construct the Reynolds force vector (RFV). This quantity was shown to produce less error propagation, significantly enhancing the accuracy of the solution of Reynolds-averaged Navier–Stokes (RANS) equations, which were shown to be ill-conditioned. This fact motivates the use of the RFV as a target for ML techniques. Invariance is enforced by means of a procedure where vectorial and tensorial quantities are represented using an intrinsic local coordinate system associated with the eigenvectors of the rate-of-strain tensor of the baseline solution. The proposed hybrid method employs an implicit procedure for the linear part of the RFV. This structure is applied to two geometries analyzed in the literature, namely the square duct and the periodic hill flows, using direct numerical simulation (DNS) as the data source for learning and κ – ϵ RANS simulations as the baseline solution. A random-forest (RF) approach is employed for the machine learning process. When compared with other approaches, the new hybrid method shows lower error propagation from the predictions of the ML approach for both geometries, illustrating the potential of this model strategy. [ABSTRACT FROM AUTHOR]

Analogical proportions are 4-ary relations that read "A is to B as C is to D". Recent works have highlighted the fact that such relations can support a specific form of inference, called analogical inference. This inference mechanism was empirically proved to be efficient in several reasoning and classification tasks. In the latter case, it relies on the notion of analogy preservation. In this paper, we explore this relation between formal models of analogy and the corresponding classes of analogy preserving functions, and we establish a Galois theory of analogical classifiers. We illustrate the usefulness of this Galois framework over Boolean domains, and we explicitly determine the closed sets of analogical classifiers, i.e., classifiers that are compatible with the analogical inference, for each pair of Boolean analogies. [ABSTRACT FROM AUTHOR]

Distribution System Operators (DSOs) and Aggregators benefit from novel Energy Generation Forecasting (EGF) approaches. Improved forecasting accuracy may make it easier to deal with energy imbalances between production and consumption. It also aids operations such as Demand Response (DR) management in Smart Grid architecture. This work aims to develop and test a new solution for EGF. It combines various methodologies running EGF tests on historical data from buildings. The experimentation yields different data resolutions (15 min, one hour, one day, etc.) while reporting accuracy errors. The optimal forecasting technique should be relevant to a variety of forecasting applications in a trial-and-error manner, while utilizing different forecasting strategies, ensemble approaches, and algorithms. The final forecasting evaluation incorporates performance metrics such as coefficient of determination ( R 2 ), Mean Absolute Error (MAE), Mean Squared Error (MSE) and Root Mean Squared Error (RMSE), presenting a comparative analysis of results. [ABSTRACT FROM AUTHOR]

Dengue is a mosquito-borne pandemic viral infection, which transmits to humans from Female Aedes albopictis or Aedes agypti mosquitoes. It progressively deteriorates the health of infected individuals and poses a high threat of human morbidity and mortality. This paper proposes an intelligent healthcare system which identifies, monitors, and alerts dengue virus (DeV) infected individuals and other stakeholders in real-time and control the DeV infection outbreak using cloud computing, internet of things and fog computing paradigms. The proposed system uses Naive Bayesian Network (NBN) for diagnosing the possibly DeV infected individuals and generating real-time alerts for suggesting and alerting the concerned stakeholders for taking on-time necessary actions at the fog subsystem. The proposed system also uses Social Network Analysis at the cloud subsystem, to provide Global Positioning Systems (GPS)-based global risk assessment of the DeV infection on Google Maps (Google-based web map service) and control DeV infection outbreak. The analysis of the experimental results acknowledges the efficiency of the NBN-based DeV infection diagnosis, alert generation, and GPS-based risk assessment functionality, of the proposed system, via various statistical measures and experimental approaches. [ABSTRACT FROM AUTHOR]

Based on the virtualization technology and pushed by the softwarization paradigm and the actual demand for services and resources, commercial cloud data centers know an unprecedented expansion. The systematic presence of software and the services generated enabled the development of the already dense application expenses. This causes, not only a cost explosion, especially when proprietary solutions protected by licenses are in hand, but also, represents a critical need in terms of software asset and resource management at the SaaS level. In addition to these costs, inefficient resource utilization, and the resulting energy represent an important part of the operational expenditure of data centers and are still a hot topic despite the consolidation initiatives put in place. The main objective of the consolidation service is to maximize resource exploitation while minimizing energy consumption and costs, among others. Even so, we have noticed that the reported literature doesn't treat license management in the cloud environment as a whole, especially, from the resource management perspective and the overwhelming majority of the consolidation work focuses on resource optimization at the IaaS level. Therefore, we propose a reinforcement learning-based scheme that allows efficient use of resources and optimizes costs, energy consumption, and resource wastage, while remaining compliant. The experimental results show that our intelligent consolidator outperforms the baseline approaches according to the evaluation metrics used regardless of the resource heterogeneity and the data center dimensionality. [ABSTRACT FROM AUTHOR]

Using polarity, we give an outer polyhedral approximation for the epigraph of set functions. For a submodular function, we prove that the corresponding polar relaxation is exact; hence, it is equivalent to the Lovász extension. The polar approach provides an alternative proof for the convex hull description of the epigraph of a submodular function. Computational experiments show that the inequalities from outer approximations can be effective as cutting planes for solving submodular as well as non-submodular set function minimization problems. [ABSTRACT FROM AUTHOR]

In this article we study the stochastic gradient descent (SGD) optimization method in the training of fully connected feedforward artificial neural networks with ReLU activation. The main result of this work proves that the risk of the SGD process converges to zero if the target function under consideration is constant. In the established convergence result the considered artificial neural networks consist of one input layer, one hidden layer, and one output layer (with d ∈ N neurons on the input layer, H ∈ N neurons on the hidden layer, and one neuron on the output layer). The learning rates of the SGD process are assumed to be sufficiently small, and the input data used in the SGD process to train the artificial neural networks is assumed to be independent and identically distributed. [ABSTRACT FROM AUTHOR]

Abundant research has been conducted recently on recommendation systems. A recommendation system is a subfield of information retrieval and machine learning that aims to identify the value of objects and information for everyone. For example, recommendation systems on web services learn the latent preferences of users based on their behavioral history and display content that the system considers as s user favorite. Specifically, recommendation systems for web services have two main requirements: (1) use the embedding of users and items for predictions and (2) use implicit feedback data that does not require users' active actions when learning. Recently, a method called collaborative metric learning (CML) has been developed to satisfy the first requirement. However, this method does not address noisy label issues caused by implicit feedback data in the second requirement. This study proposes a comprehensive and effective method to deal with noise in CML. The experimental results show that the proposed method significantly improves the performance of the two requirements compared with existing methods. [ABSTRACT FROM AUTHOR]

With wind power providing an increasing amount of electricity worldwide, the quantification of its spatio-temporal variations and the related uncertainty is crucial for energy planners and policy-makers. Here, we propose a methodological framework which (1) uses machine learning to reconstruct a spatio-temporal field of wind speed on a regular grid from spatially irregularly distributed measurements and (2) transforms the wind speed to wind power estimates. Estimates of both model and prediction uncertainties, and of their propagation after transforming wind speed to power, are provided without any assumptions on data distributions. The methodology is applied to study hourly wind power potential on a grid of 250 × 250 m 2 for turbines of 100 m hub height in Switzerland, generating the first dataset of its type for the country. We show that the average annual power generation per turbine is 4.4 GWh. Results suggest that around 12,000 wind turbines could be installed on all 19,617 km 2 of available area in Switzerland resulting in a maximum technical wind potential of 53 TWh. To achieve the Swiss expansion goals of wind power for 2050, around 1000 turbines would be sufficient, corresponding to only 8% of the maximum estimated potential. [ABSTRACT FROM AUTHOR]

The mapping and analysis of scientific knowledge makes it possible to identify the dynamics and/or growth of a particular field of research or to support strategic decisions related to different research entities, based on bibliometric and/or scientometric indicators. However, with the exponential growth of scientific production, a systematic and data-oriented approach to the analysis of this large set of productions becomes increasingly essential. Thus, in this work, a data-oriented methodology was proposed, combining Data Analysis, Machine Learning and Complex Network Analysis techniques, and Data Version Control (DVC) tool, for the extraction of implicit knowledge in scientific production bases. In addition, the approach was validated through a case study in a COVID-19 manuscripts dataset, which had 199,895 articles published on arXiv, bioRxiv, medRxiv, PubMed and Scopus databases. The results suggest the feasibility of the proposed methodology, indicating the most active countries and the most explored themes in each period of the pandemic. Therefore, this study has the potential to instrument and expand strategic decisions by the scientific community, aiming at extracting knowledge that supports the fight against the COVID-19 pandemic. [ABSTRACT FROM AUTHOR]

This article is concerned with the approximation and expressive powers of deep neural networks. This is an active research area currently producing many interesting papers. The results most commonly found in the literature prove that neural networks approximate functions with classical smoothness to the same accuracy as classical linear methods of approximation, e.g., approximation by polynomials or by piecewise polynomials on prescribed partitions. However, approximation by neural networks depending on n parameters is a form of nonlinear approximation and as such should be compared with other nonlinear methods such as variable knot splines or n-term approximation from dictionaries. The performance of neural networks in targeted applications such as machine learning indicate that they actually possess even greater approximation power than these traditional methods of nonlinear approximation. The main results of this article prove that this is indeed the case. This is done by exhibiting large classes of functions which can be efficiently captured by neural networks where classical nonlinear methods fall short of the task. The present article purposefully limits itself to studying the approximation of univariate functions by ReLU networks. Many generalizations to functions of several variables and other activation functions can be envisioned. However, even in this simplest of settings considered here, a theory that completely quantifies the approximation power of neural networks is still lacking. [ABSTRACT FROM AUTHOR]

We develop a methodology for classifying defects based on their morphology and induced mechanical response. The proposed approach is fairly general and relies on morphological operators (Angulo and Meyer in 9th international symposium on mathematical morphology and its applications to signal and image processing, pp. 226-237, 2009) and spherical harmonic decomposition as a way to characterize the geometry of the pores, and on the Grassman distance evaluated on FFT-based computations (Willot in C. R., Méc. 343(3):232–245, 2015), for the predicted elastic response. We implement and detail our approach on a set of trapped gas pores observed in X-ray tomography of welded joints, that significantly alter the mechanical reliability of these materials (Lacourt et al. in Int. J. Numer. Methods Eng. 121(11):2581–2599, 2020). The space of morphological and mechanical responses is first partitioned into clusters using the "k-medoids" criterion and associated distance functions. Second, we use multiple-layer perceptron neural networks to associate a defect and corresponding morphological representation to its mechanical response. It is found that the method provides accurate mechanical predictions if the training data contains a sufficient number of defects representing each mechanical class. To do so, we supplement the original set of defects by data augmentation techniques. Artificially-generated pore shapes are obtained using the spherical harmonic decomposition and a singular value decomposition performed on the pores signed distance transform. We discuss possible applications of the present method, and how medoids and their associated mechanical response may be used to provide a natural basis for reduced-order models and hyper-reduction techniques, in which the mechanical effects of defects and structures are decorrelated (Ryckelynck et al. in C. R., Méc. 348(10–11):911–935, 2020). [ABSTRACT FROM AUTHOR]

This paper focuses on how to approximate traveling wave solutions for various kinds of partial differential equations via artificial neural networks. A traveling wave solution is hard to obtain with traditional numerical methods when the corresponding wave speed is unknown in advance. We propose a novel method to approximate both the traveling wave solution and the unknown wave speed via a neural network and an additional free parameter. We proved that under a mild assumption, the neural network solution converges to the analytic solution and the free parameter accurately approximates the wave speed as the corresponding loss tends to zero for the Keller–Segel equation. We also demonstrate in the experiments that reducing loss through training assures an accurate approximation of the traveling wave solution and the wave speed for the Keller–Segel equation, the Allen–Cahn model with relaxation, and the Lotka–Volterra competition model. [ABSTRACT FROM AUTHOR]

This paper provides guidelines for an effective artificial neural networks (ANNs) design to aid stabilized finite element schemes. In particular, ANNs are used to estimate the stabilization parameter of the streamline upwind Petrov–Galerkin (SUPG) stabilization scheme for singularly perturbed problems. The effect of the artificial neural network (ANN) hyper-parameters on the accuracy of ANNs is found by performing a global sensitivity analysis. First, a Gaussian process regression metamodel of the artificial neural networks is obtained. Next, analysis of variance is performed to obtain Sobol' indices. The total-order Sobol' indices identify the hyper-parameters having the maximum effect on the accuracy of the ANNs. Furthermore, the best-performing and the worst-performing networks are identified among the candidate ANNs. Our findings are validated with the help of one-dimensional test cases in the advection-dominated flow regime. This study provides insights into hyper-parameters' effect and consequently aids in building effective ANN models for applications involving nonlinear regression, including estimation of SUPG stabilization parameters. [ABSTRACT FROM AUTHOR]

Modeling a combinatorial problem is a hard and error-prone task requiring significant expertise. Constraint acquisition methods attempt to automate this process by learning constraints from examples of solutions and (usually) non-solutions. Active methods query an oracle while passive methods do not. We propose a known but not widely-used application of machine learning to constraint acquisition: training a classifier to discriminate between solutions and non-solutions, then deriving a constraint model from the trained classifier. We discuss a wide range of possible new acquisition methods with useful properties inherited from classifiers. We also show the potential of this approach using a Naive Bayes classifier, obtaining a new passive acquisition algorithm that is considerably faster than existing methods, scalable to large constraint sets, and robust under errors. [ABSTRACT FROM AUTHOR]

Assigning appropriate reviewers to a manuscript from a pool of candidate reviewers is a common challenge in the academic community. Current word- and semantic-based approaches treat the reviewer assignment problem (RAP) as an information retrieval problem but do not take into account two constraints of the RAP: incompleteness of the reviewer data and interference from nonmanuscript-related papers. In this paper, a word and semantic-based iterative model (WSIM) is proposed to account for the constraints of the RAP by improving the similarity calculations between reviewers and manuscripts. First, we use the improved language model and topic model to extract word features and semantic features to represent reviewers and manuscripts. Second, we use a similarity metric based on the normalized discounted cumulative gain (NDCG) to measure semantic similarity. This metric ignores the probability value (quantitative exact value) of the topic and considers only the ranking (qualitative relevance), thus reducing overfitting to incomplete reviewer data. Finally, we use an iterative model to reduce the interference from nonmanuscript-related papers in the reviewer data. This approach considers the similarity between the manuscript and each of the reviewer's papers. We evaluate the proposed WSIM on two real datasets and compare its performance to that of seven existing methods. The experimental results show that the WSIM improves the recommendation accuracy by at least 2.5% on the top 20. [ABSTRACT FROM AUTHOR]

Sensor placement is an optimisation problem that has recently gained great relevance. In order to achieve accurate online updates of a predictive model, sensors are used to provide observations. When sensor location is optimally selected, the predictive model can greatly reduce its internal errors. A greedy-selection algorithm is used for locating these optimal spatial locations from a numerical embedded space. A novel architecture for solving this big data problem is proposed, relying on a variational Gaussian process. The generalisation of the model is further improved via the preconditioning of its inputs: Masked Autoregressive Flows are implemented to learn nonlinear, invertible transformations of the conditionally modelled spatial features. Finally, a global optimisation strategy extending the Mutual Information-based optimisation and fine-tuning of the selected optimal location is proposed. The methodology is parallelised to speed up the computational time, making these tools very fast despite the high complexity associated with both spatial modelling and placement tasks. The model is applied to a real three-dimensional test case considering a room within the Clarence Centre building located in Elephant and Castle, London, UK. [ABSTRACT FROM AUTHOR]

Spiking neural networks (SNNs) with a lattice architecture are introduced in this work, combining several desirable properties of SNNs and self-organized maps (SOMs). Networks are trained with biologically motivated, unsupervised learning rules to obtain a self-organized grid of filters via cooperative and competitive excitatory-inhibitory interactions. Several inhibition strategies are developed and tested, such as (i) incrementally increasing inhibition level over the course of network training, and (ii) switching the inhibition level from low to high (two-level) after an initial training segment. During the labeling phase, the spiking activity generated by data with known labels is used to assign neurons to categories of data, which are then used to evaluate the network's classification ability on a held-out set of test data. Several biologically plausible evaluation rules are proposed and compared, including a population-level confidence rating, and an n-gram inspired method. The effectiveness of the proposed self-organized learning mechanism is tested using the MNIST benchmark dataset, as well as using images produced by playing the Atari Breakout game. [ABSTRACT FROM AUTHOR]

Community detection plays an important role in the analysis of complex networks. However, overlapping community detection in real networks is still a challenge. To address the problems of pre-input parameters and label redundancy, an improved label propagation algorithm (ILPA) that adopts a method based on the influence factor is proposed in this paper. Theoretical analysis and experimental results on both synthetic and real datasets show that the ILPA detects that the overlapping community has higher accuracy compared to other existing methods. [ABSTRACT FROM AUTHOR]

Time series averaging is an important subroutine for several time series data mining tasks. The most successful approaches formulate the problem of time series averaging as an optimization problem based on the dynamic time warping (DTW) distance. The existence of an optimal solution, called sample mean, is an open problem for more than four decades. Its existence is a necessary prerequisite to formulate exact algorithms, to derive complexity results, and to study statistical consistency. In this article, we propose sufficient conditions for the existence of a sample mean. A key result for deriving the proposed sufficient conditions is the Reduction Theorem that provides an upper bound for the minimum length of a sample mean. [ABSTRACT FROM AUTHOR]

We consider the binary classification problem when data are large and subject to unknown but bounded uncertainties. We address the problem by formulating the nonlinear support vector machine training problem with robust optimization. To do so, we analyze and propose two bounding schemes for uncertainties associated to random approximate features in low dimensional spaces. The proposed bound calculations are based on Random Fourier Features and the Nyström methods. Numerical experiments are conducted to illustrate the benefit of the technique. We also emphasize the decomposable structure of the proposed robust nonlinear formulation that allows the use of efficient stochastic approximation techniques when datasets are large. [ABSTRACT FROM AUTHOR]

This paper proposes an algorithm to reason on constraints expressed in terms of polynomials with integer coefficients whose variables take values from finite subsets of the integers. The proposed algorithm assumes that an initial approximation of the domains of variables is available in terms of a bounding box, and it recursively subdivides the box into disjoint boxes until a termination condition is met. The algorithm includes three termination conditions that allow using it for three related reasoning tasks: constraint satisfaction, enumeration of solutions, and hyper-arc consistency enforcement. Considered termination conditions are based on suitable lower and upper bounds for polynomial functions over boxes that are determined using new results proved in the paper. The algorithm is particularly appropriate to reason on high-degree polynomial constraints because the proposed method to determine lower and upper bounds can outperform alternative methods when high-degree polynomials in a moderate number of variables are considered. [ABSTRACT FROM AUTHOR]

Due to the development of academic, more and more attentions are paid to citation recommendation. To solve the citation recommendation problem, researchers begin to focus on the network representation, because it fuses semantic information and structural information well. It is a big challenge that how to map articles in a heterogeneous information network into a low-dimensional space while preserving the potential associations between articles. We propose a novel citation recommendation algorithm based on citation tendency, named CIRec which learns more about the potential relationship of articles in the process of network embedding. Citation tendency means if an article can be selected as a reference, it probability satisfies some kinds of conditions. In our algorithm, five weight matrices which represent the probability of entity-to-entity migration based on citation tendency are defined to build weighted heterogeneous network first. Second, we design a biased random walk procedure which efficiently explores articles' characteristics and citations information. Finally, the skip-gram model is used to learn the neighborhood relationship of the nodes in the walk sequence and map the nodes to the vector space. Comparing with existing state-of-the-art technique, experiment results show that our algorithm CIRec has better recall, precision, NDCG on AAN and DBLP dataset. [ABSTRACT FROM AUTHOR]

In this paper, a global asymptotic stability condition for Long Short-Term Memory neural networks is presented. Since this condition is formulated in terms of the networks' weight matrices and biases that are essentially control variables, the same condition can be viewed as a way to globally asymptotically stabilize these networks. The condition and how to compute numerical values for the weight matrices and biases are illustrated by a number of numerical examples. [ABSTRACT FROM AUTHOR]

Network representation learning (NRL) enables the application of machine learning tasks such as classification, prediction and recommendation to networks. Apart from their graph structure, networks are often associated with diverse information in the form of attributes. Most NRL methods have focused just on structural information, and separately apply a traditional representation learning on attributes. When multiple sources of information are available, using a combination of them may be beneficial as they complement each other in generating accurate contexts; moreover, their combined use may be fundamental when the information sources are sparse. The learning methods should thus preserve both the structural and attribute aspects. In this paper, we investigate how attributes can be modeled, and subsequently used along with structural information in learning the representation. We introduce the GAT2VEC framework that uses structural information to generate structural contexts, attributes to generate attribute contexts, and employs a shallow neural network model to learn a joint representation from them. We evaluate our proposed method against state-of-the-art baselines, using real-world datasets on vertex classification (multi-class and multi-label), link-prediction, and visualization tasks. The experiments show that GAT2VEC is effective in exploiting multiple sources of information, thus learning accurate representations and outperforming the state-of-the-art in the aforementioned tasks. Finally, we perform query tasks on learned representation and show how the qualitative analysis of results has better performance as well. [ABSTRACT FROM AUTHOR]

Reviewer recommendation problem in the research field usually refers to invite experts to comment on the quality of papers, proposals, etc. How to effectively and accurately recommend reviewers for the submitted papers and proposals is a meaningful and still tough task. At present, many unsupervised recommendation methods have been researched to solve this task. In this paper, a novel classification method named Word Mover’s Distance-Constructive Covering Algorithm (WMD-CCA, for short) is proposed to solve the reviewer recommendation problem as a classification issue. A submission or a reviewer is described by some tags, such as keywords, research interests, and so on. First, the submission or the reviewer is represented as some vectors by a word embedding method. That is to say, each tag describing a submission or a reviewer is represented as a vector. Second, the Word Mover’s Distance (WMD, for short) method is used to measure the minimum distances between submissions and reviewers. Actually, the papers usually have research field information, and utilizing them well might improve the reviewer recommendation accuracy. So finally, the reviewer recommendation task is transformed into a classification problem which is solved by a supervised learning method- Constructive Covering Algorithm (CCA, for short). Comparative experiments are conducted with 4 public datasets and a synthetic dataset from Baidu Scholar, which show that the proposed method WMD-CCA effectively solves the reviewer recommendation task as a classification issue and improves the recommendation accuracy. [ABSTRACT FROM AUTHOR]

Control by partition refers to situations where an election chair seeks to influence the outcome of an election by partitioning either the candidates or the voters into two groups, thus creating two first-round subelections that determine who will take part in a final round. The model of partition-of-voters control attacks is remotely related to “gerrymandering” (maliciously resizing election districts). While the complexity of control by partition has been studied thoroughly for many voting systems, there are no such results known for the important veto voting system. We settle the complexity of control by partition for veto in a broad variety of models. In addition, by giving a counterexample we observe that a reduction from the literature (Chen et al. 2015) showing the parameterized complexity of control by adding candidates to plurality elections, parameterized by the number of voters, is technically flawed, and we show how this reduction can be adapted to make it correct. [ABSTRACT FROM AUTHOR]

This paper describes an algorithm to enforce hyper-arc consistency of polynomial constraints defined over finite domains. First, the paper describes the language of so called polynomial constraints over finite domains, and it introduces a canonical form for such constraints. Then, the canonical form is used to transform the problem of testing the satisfiability of a constraint in a box into the problem of studying the sign of a related polynomial function in the same box, a problem which is effectively solved by using the modified Bernstein form of polynomials. The modified Bernstein form of polynomials is briefly discussed, and the proposed hyper-arc consistency algorithm is finally detailed. The proposed algorithm is a subdivision procedure which, starting from an initial approximation of the domains of variables, removes values from domains to enforce hyper-arc consistency. [ABSTRACT FROM AUTHOR]

Web Service Composition aims to select and aggregate many web services to generate a workflow. The workflow contains many tasks and for each task there are many web services to choose from. The challenge is to select the best combination of web services that achieve the user requirements. This problem is called Web Service Selection (WSS). In this work, we improve on the Artificial Bee Colony Algorithm to make it more suitable for the WSS problem. Our proposed enhancement controls the exploitation and exploration strategies in such a way that encourages exploration at early stages and exploitation at later stages. Our experiments indicate that our algorithm finds better solutions and reduces the execution time compared with other algorithms. [ABSTRACT FROM AUTHOR]

Submodularity defines a general framework rich of important theoretical properties while accommodating numerous applications. Although the notion has been present in the literature of Combinatorial Optimization for several decades, it has been overlooked in the analysis of global constraints. The current work illustrates the potential of submodularity as a powerful tool for such an analysis. In particular, we show that the cumulative constraint, when all tasks are identical, has the submodular/supermodular representation property, i.e., it can be represented by a submodular/supermodular system of linear inequalities. Motivated by that representation, we show that the system of any two (global) constraints not necessarily of the same type, each bearing the above-mentioned property, has an integral relaxation given by the conjunction of the linear inequalities representing each individual constraint. This result is obtained through the use of the celebrated polymatroid intersection theorem. [ABSTRACT FROM AUTHOR]

Decision trees (DT) are highly famous in machine learning and usually acquire state-of-the-art performance. Despite that, well-known variants like CART, ID3, random forest, and boosted trees miss a probabilistic version that encodes prior assumptions about tree structures and shares statistical strength between node parameters. Existing work on Bayesian DT depends on Markov Chain Monte Carlo (MCMC), which can be computationally slow, especially on high dimensional data and expensive proposals. In this study, we propose a method to parallelise a single MCMC DT chain on an average laptop or personal computer that enables us to reduce its run-time through multi-core processing while the results are statistically identical to conventional sequential implementation. We also calculate the theoretical and practical reduction in run time, which can be obtained utilising our method on multi-processor architectures. Experiments showed that we could achieve 18 times faster running time provided that the serial and the parallel implementation are statistically identical. [ABSTRACT FROM AUTHOR]

Recent years have seen an increased interest in crowdsourcing as a way of obtaining information from a potentially large group of workers at a reduced cost. The crowdsourcing process, as we consider in this paper, is as follows: a requester hires a number of workers to work on a set of similar tasks. After completing the tasks, each worker reports back outputs. The requester then aggregates the reported outputs to obtain aggregate outputs. A crucial question that arises during this process is: how many crowd workers should a requester hire? In this paper, we investigate from an empirical perspective the optimal number of workers a requester should hire when crowdsourcing tasks, with a particular focus on the crowdsourcing platform Amazon Mechanical Turk. Specifically, we report the results of three studies involving different tasks and payment schemes. We find that both the expected error in the aggregate outputs as well as the risk of a poor combination of workers decrease as the number of workers increases. Surprisingly, we find that the optimal number of workers a requester should hire for each task is around 10 to 11, no matter the underlying task and payment scheme. To derive such a result, we employ a principled analysis based on bootstrapping and segmented linear regression. Besides the above result, we also find that overall top-performing workers are more consistent across multiple tasks than other workers. Our results thus contribute to a better understanding of, and provide new insights into, how to design more effective crowdsourcing processes. [ABSTRACT FROM AUTHOR]

The notion of contact algebra is one of the main tools in the region based theory of space. It is an extension of Boolean algebra with an additional relation C called contact. The elements of the Boolean algebra are considered as formal representations of spatial regions as analogues of physical bodies and Boolean operations are considered as operations for constructing new regions from given ones and also to define some mereological relations between regions as part-of, overlap and underlap. The contact relation is one of the basic mereotopological relations between regions expressing some topological nature. It is used also to define some other important mereotopological relations like non-tangential inclusion, dual contact, external contact and others. Most of these definitions are given by means of the operation of Boolean complementation. There are, however, some problems related to the motivation of the operation of Boolean complementation. In order to avoid these problems we propose a generalization of the notion of contact algebra by dropping the operation of complement and replacing the Boolean part of the definition by that of a distributive lattice. First steps in this direction were made in (Düntsch et al. Lect. Notes Comput. Sci. 4136, 135-147, 2006, Düntsch et al. J. Log. Algebraic Program. 76, 18-34, 2008) presenting the notion of distributive contact lattice based on the contact relation as the only mereotopological relation. In this paper we consider as non-definable primitives the relations of contact, nontangential inclusion and dual contact, extending considerably the language of distributive contact lattices. Part I of the paper is devoted to a suitable axiomatization of the new language called extended distributive contact lattice (EDC-lattice) by means of universal first-order axioms true in all contact algebras. EDC-lattices may be considered also as an algebraic tool for a certain subarea of mereotopology, called in this paper distributive mereotopology. The main result of Part I of the paper is a representation theorem, stating that each EDC-lattice can be isomorphically embedded into a contact algebra, showing in this way that the presented axiomatization preserves the meaning of mereotopological relations without considering Boolean complementation. Part II of the paper is devoted to topological representation theory of EDC-lattices, transferring into the distributive case important results from the topological representation theory of contact algebras. It is shown that under minor additional assumptions on distributive lattices as extensionality of the definable relations of overlap or underlap one can preserve the good topological interpretations of regions as regular closed or regular open sets in topological space. [ABSTRACT FROM AUTHOR]

With advances in technology, frequent pattern mining has been used widely in our daily lives. By using this technology, one can obtain interesting or useful information that would help one make decisions and apply judgment. For example, marketplace managers mine transaction data to obtain information that can help improve services, understand customer buying habits, determine a suitable scheme for placement of goods to increase profits, or for medical and biotechnology applications. However, the rate at which data is generated is very rapid, leading to problems caused by Big Data. Therefore, many researchers have studied distributed, parallel and cloud computing technology to select the best among them. However, data mining uses multiple computing nodes, which requires the transmission of a considerable amount of data in a network environment. The available network bandwidth is limited when many different tasks are being transmitted at the same time and many servers are working in the same network segment. This results in poor transmission, causing severe transfer delay, either internal or external to the network. Thus, we propose the fast and distributed mining algorithm for discovering frequent patterns in congested networks (FDMCN) algorithm, which is based on CARM. The main purpose is to reduce FP-tree transmission such that only a portion of the information is required for mining using computing nodes. The results of empirical evaluation under various simulation conditions show that the proposed method FDMCN delivers excellent performance in terms of execution efficiency and scalability when compared with the PSWS algorithm. [ABSTRACT FROM AUTHOR]