Representative Algorithms Research Articles

Variable seismic waveforms representation: Weak-supervised learning based seismic horizon picking

Seismic horizon picking via deep learning models have been advanced rapidly and proven popular. However, the prediction result is highly depended on the quality of the train set and the manually interpreted train set is very difficult to obtain. To guarantee a reasonable prediction result, we usually need manually interpret 20%–30% seismic data to generate the train set for the convolutional neural network (CNN) based seismic horizon picking methods, which is very time consuming. Besides, the performance of the deep learning based seismic horizon picking methods are also greatly influenced by the architecture of the selected CNN model. The core theory of using CNN for seismic horizon picking is recognizing different waveform patterns and classifying them with the corresponding labels. In this work, we first develop the variable waveform representation (VWR) algorithm to simulate the waveform features and then produce a large number of seismic traces as the train data. The VWR can deal with the task of the train set generation by only using very few seismic traces. Afterward, we propose a novel CNN model by integrating the holistically-nested module and the SegNet model for solving seismic horizon picking, which is termed as the HSegNet model. It should be noted that the holistically-nested module is able to recognize the holistic image and learn the multiple features, which is efficient for the waveform pattern recognition. To demonstrate the efficiency and robustness of our proposed HSegNet model, we apply it to field data and compare with the widely used U-Net model. The experiment results show that our proposed VWR-HSegNet can achieve the prediction result with 99% test accuracy by only manually interpreting 0.13% seismic traces. Whereas, the U-Net model can only get 91% test accuracy prediction result but using 30% manually interpreted seismic traces as the train set.

StaSiS-Net: A stacked and siamese disparity estimation network for depth reconstruction in modern 3D laparoscopy.

Developing accurate and real-time algorithms for a non-invasive three-dimensional representation and reconstruction of internal patient structures is one of the main research fields in computer-assisted surgery and endoscopy. Mono and stereo endoscopic images of soft tissues are converted into a three-dimensional representation by the estimation of depth maps. However, automatic, detailed, accurate and robust depth map estimation is a challenging problem that, in the stereo setting, is strictly dependent on a robust estimate of the disparity map. Many traditional algorithms are often inefficient or not accurate. In this work, novel self-supervised stacked and Siamese encoder/decoder neural networks are proposed to compute accurate disparity maps for 3D laparoscopy depth estimation. These networks run in real-time on standard GPU-equipped desktop computers and the outputs may be used for depth map estimation using the a known camera calibration. We compare performance on three different public datasets and on a new challenging simulated dataset and our solutions outperform state-of-the-art mono and stereo depth estimation methods. Extensive robustness and sensitivity analyses on more than 30000 frames has been performed. This work leads to important improvements in mono and stereo real-time depth map estimation of soft tissues and organs with a very low average mean absolute disparity reconstruction error with respect to ground truth.

Time-Weighted Kernel-Sparse-Representation-Based Real-Time Nonlinear Multimode Process Monitoring

Real-time nonlinear multimode process monitoring of actual industrial systems has attracted increasing attention recently. In this article, the time-weighed kernel sparse representation (TWKSR) method is proposed to partition the mode of the training dataset by introducing the time-series-dependent characteristics into the kernel sparse representation algorithm. The alternating direction method of multipliers is utilized to solve the optimization problem of the proposed TWKSR method. Then, the representative samples from each identified mode are selected to update the dictionary matrix. Based on the updated dictionary matrix, the sparse coefficient is used for online mode identification, and the reconstruction error is utilized for fault detection. Finally, a numerical simulation case and the wastewater treatment process example verify the effectiveness of the proposed method.

Performance analysis on dictionary learning and sparse representation algorithms

Performance analysis on dictionary learning and sparse representation algorithms

Cancer MiRNA biomarker classification based on Improved Generative Adversarial Network optimized with Mayfly Optimization Algorithm

Nowadays, cancer diagnosis becomes a paradigm shift by incorporating molecular biomarkers as part of a routine diagnostic panel. Ranges of molecular changes include DNA, RNA, micro RNA (miRNAs) and proteins. In recent years, deep learning based methods have been more inspired to health researcher’s regarding the performance of cancer diagnosis. The application of deep learning-based approach gradually becomes clearer in classification accuracy for a problem that separates data related to cancer survival. In this manuscript, an Improved Generative Adversarial Network optimized with Mayfly Optimization Algorithm is proposed to overcome the super class issues. Improved Generative Adversarial Network is the combination of deep convolutional generative adversarial network (DCG) and modified convolutional neural network (MCNN); hence it is called DCG-MCNN. Initially, the DCG is used to balance the dataset by creating more samples in the training dataset. Based on the training dataset, cancer miRNA biomarker classification is improved with the help of modified CNN diagnosis model. The proposed method is activated in python, moreover, its efficiency is analyzed with Cancer Genome Atlas dataset. Here, performance metrics, viz accuracy, sensitivity, specificity, precision, F-measure balanced error rate are calculated. The experimental results of the proposed method shows higher accuracy 99.26%, higher sensitivity 95.23%, higher specificity 92.56% compared with the existing methods, like Validation of miRNAs as breast cancer biomarkers with a machine learning approach (CMiRNA-BC-RF-SVM), Cancer miRNA biomarkers classification using a new representation algorithm and evolutionary deep learning (CMiRNA-BC-CNN) and multi-omics data using graph convolutional networks allowing patient classification and biomarker identification (CMiRNA-BC-GCNN).

Experimental Study of Algorithms for Minimization of Binary Decision Diagrams using Algebraic Representations of Cofactors

BDD (Binary Decision Diagram) is used for technology-independent optimization, performed as the first stage in the synthesis of logic circuits in the design of ASIC (application-specific integrated circuit). BDD is an acyclic graph defining a Boolean function or a system of Boolean functions. Each vertex of this graph is associated with the complete or reduced Shannon expansion formula. Binary decision diagrams with mutually inverse subfunctions (cofac-tors) are considered. We have developed algorithms for finding algebraic representations of cofactors of the same BDD level in the form of a disjunction or conjunction of other inverse or non-inverse cofactors of the same BDD level. The algorithms make it possible to reduce the number of literals by replacing the Shannon expansion formulas with simpler logical formulas and to reduce the number of literals in the description of a system of Boolean functions. We propose to use the developed algorithms for an additional logical optimization of the constructed BDD representations of systems of Boolean functions. Experimental results of the application of the corresponding programs in the synthesis of logic circuits in the design library of custom VLSI CMOS circuits are presented.

Learning multiplication-free linear transformations

In this paper, we propose several dictionary learning algorithms for sparse representations that also impose specific structures on the learned dictionaries such that they have low coding complexity and are numerically efficient to use: reduced number of addition/multiplications and even avoid multiplications altogether. We base our work on factorizations of the dictionaries in highly structured basic building blocks (binary orthonormal, scaling, and shear transformations) for which we can write closed-form solutions to the optimization problems that we consider. We show the effectiveness of our methods on image data where we compare against well-known numerically efficient transforms such as fast Fourier, fast discrete cosine transforms, and fast structured dictionaries.

Quantum matching pursuit: A quantum algorithm for sparse representations

Representing signals with sparse vectors has a wide range of applications that range from image and video coding to shape representation and health monitoring. In many applications with real-time requirements, or that deal with high-dimensional signals, the computational complexity of the encoder that finds the sparse representation plays an important role. Quantum computing has recently shown promising speed-ups in many representation learning tasks. In this work, we propose a quantum version of the well-known matching pursuit algorithm. Assuming the availability of a fault-tolerant quantum random access memory, our quantum matching pursuit lowers the complexity of its classical counterpart of a polynomial factor, at the cost of some error in the computation of the inner products, enabling the computation of sparse representation of high-dimensional signals. Besides proving the computational complexity of our new algorithm, we provide numerical experiments that show that its error is negligible in practice. This work opens the path to further research on quantum algorithms for finding sparse representations, showing suitable quantum computing applications in signal processing.

Algorithms for the Minimal Rational Fraction Representation of Sequences Revisited

Algorithms for the Minimal Rational Fraction Representation of Sequences Revisited

Decision support system for adaptive sourcing and inventory management in small- and medium-sized enterprises

• Proposes a data-driven approach in sourcing and inventory management for SMEs. • Exploits Observational and empirical data to generate decision support solutions. • Promote the adoption and initiation of artificial intelligence in SMEs. • Promising implementation results in a real-world case study. Elevated business uncertainties and competition over recent years have caused changes to the data-driven supply chain management of sourcing and inventories across industries. However, only large-sized enterprises have the resources to harness data for aiding their decision-making and planning. By contrast, small- and medium-sized enterprises (SMEs) commonly have limited resources and knowledge, which affects their ability to collect and utilize data. Thus, it is a challenge for them to implement advanced decision support tools to mitigate the effects of market uncertainties. This paper proposes a decision support system (DSS) for sourcing and inventory management, with the aims of helping SMEs compile and exploit data, and supporting their decisions under business ambiguities. The DSS was developed using a simulation-optimization approach by incorporating an artificial neural network and a genetic algorithm for problem representation and optimizing decision support solutions. The exploitation of observational and empirical data reduces the burden of data compilation obtained from unorganized data sources across SME operations. Further, uncertainty factors such as raw material demand, price, and supply lead time were considered. When implemented in a medium-sized food industry company, the DSS can provide decision support solutions that integrate the selection of recommended suppliers and optimal order quantities. It can also help decision-makers to shape their inventory management policies under uncertain raw material demands, while also considering service levels, sales promotions, lead times, and material availability from multiple suppliers. Consequently, implementation of the DSS helped to reduce the total purchased raw material costs by an average of 51.62% and reduced the order interval and on-hand inventory costs by an average of 54.24%.

Improved sparse representation algorithm for optical coherence tomography images

AbstractAn improved sparse representation algorithm is proposed for reducing speckle noise in optical coherence tomography (OCT) images. To reduce the dictionary training time, the orthogonal matching pursuit algorithm (OMP) is replaced by a piecewise orthogonal matching pursuit algorithm (ST‐OMP), and a threshold standard is set to train multiple atoms simultaneously to improve the tracking efficiency. Normal, diabetic macular edema, and fundus neovascularization membrane retina OCT images were tested using the proposed algorithm and other denoising algorithms. The results show that the proposed algorithm exhibits an excellent denoising ability. The details and edge information in the OCT images were well preserved by the proposed algorithm. Meanwhile, the image processing time was reduced by half with the proposed algorithm compared with the traditional K‐SVD algorithm.

Analysis of Poetry Style Based on Text Classification Algorithm

In order to realize the effect of intelligent poetry style analysis, this paper applies the text classification algorithm to the poetry style analysis, combines the knowledge representation algorithm to perform text classification and recognition, improves the algorithm, and applies it to the poetry style analysis model. Moreover, this paper combines intelligent algorithms to construct a poetry style analysis system, constructs the system’s functional modules, preprocesses the poetry documents in the corpus, and maps them to the vector space that can be directly processed by the computer. In addition, after constructing the system model, this paper verifies the poetry style analysis system based on the text classification algorithm through simulation experiments. From the research results, the effect of the poetry style analysis method based on the text classification algorithm proposed in this paper is very good, which meets the actual needs of poetry style analysis.

AI-aided general clinical diagnoses verified by third-parties with dynamic uncertain causality graph extended to also include classification

Artificial intelligence (AI)-aided general clinical diagnosis is helpful to primary clinicians. Machine learning approaches have problems of generalization, interpretability, etc. Dynamic Uncertain Causality Graph (DUCG) based on uncertain casual knowledge provided by clinical experts does not have these problems. This paper extends DUCG to include the representation and inference algorithm for non-causal classification relationships. As a part of general clinical diagnoses, six knowledge bases corresponding to six chief complaints (arthralgia, dyspnea, cough and expectoration, epistaxis, fever with rash and abdominal pain) were constructed through constructing subgraphs relevant to a chief complaint separately and synthesizing them together as the knowledge base of the chief complaint. A subgraph represents variables and causalities related to a single disease that may cause the chief complaint, regardless of which hospital department the disease belongs to. Verified by two groups of third-party hospitals independently, total diagnostic precisions of the six knowledge bases ranged in 96.5–100%, in which the precision for every disease was no less than 80%.

Local trend analysis method of hydrological time series based on piecewise linear representation and hypothesis test

It is important to analyze the nonstationarity of hydrological time series under the influence of climate change and human activities. Compared with previous studies, this study focuses more on local trends rather than overall trend and mean value change, and presents a local trend analysis method, namely PLRHT method, for the nonstationary analysis of hydrological time series. In the PLRHT method, the fitted local trend lines determined by the piecewise linear representation algorithm and Akaike information criterion are used as the baseline component of hydrological time series. The difference between two adjacent fitted local trend lines includes the slope and intercept differences, which jointly determine the mean value difference of two corresponding subseries. In addition, based on the slope and intercept differences, three change points of the hydrological time series (inflection point, break point and bivariate point) are defined and their significance is tested by hypothesis test based on Monte Carlo experiments. Finally, the application effect of the PLRHT method is compared with five other representative methods of non-stationarity analysis (runs test, rank sum test, Brown-Forsythe test, T test and heuristic segmentation) by using artificial and observed hydrological time series as examples. The results show that, compared with the five nonstationarity analysis methods, the PLRHT method could not only detect more accurate change points of the artificial and observed hydrological time series, but also quantitatively analyze the contributions of trends and abrupt changes to mean value variations of these hydrological time series. Besides, the local trend analysis enables a deeper causality analysis for the nonstationarity of the observed hydrological time series. Therefore, local trend analysis could help understanding the variation characteristics and driving forces of hydrological processes in the changing environment, and the PLRHT method is a feasible local trend analysis method to reveal the relationship between local mean value changes and local trends.

Slam Algorithms for Different Map Representations

The paper discusses the generalized simultaneous localization and mapping problem statement from the standpoint of the Bayesian approach and its relationship with algorithms for different map representations. The two-dimensional example describes the linearized simultaneous localization and mapping algorithm for the mobile platform in two-dimensional space.

Retracted] Automatic Classification Method of Music Genres Based on Deep Belief Network and Sparse Representation

Aiming at the problems of poor classification effect, low accuracy, and long time in the current automatic classification methods of music genres, an automatic classification method of music genres based on deep belief network and sparse representation is proposed. The music signal is preprocessed by framing, pre‐emphasis, and windowing, and the characteristic parameters of the music signal are extracted by Mel frequency cepstrum coefficient analysis. The restricted Boltzmann machine is trained layer by layer to obtain the connection weights between layers of the depth belief network model. According to the output classification, the connection weights in the model are fine‐tuned by using the error back‐propagation algorithm. Based on the deep belief network model after fine‐tuning training, the structure of the music genre classification network model is designed. Combined with the classification algorithm of sparse representation, for the training samples of sparse representation music genre, the sparse solution is obtained by using the minimum norm, the sparse representation of test vector is calculated, the category of training samples is judged, and the automatic classification of music genre is realized. The experimental results show that the music genre automatic classification effect of the proposed method is better, the classification accuracy rate is higher, and the classification time can be effectively shortened.

NOMP: A New Sparse Solution to Enhance the SSIM Levels of OMP-Based Encoded Images

This work comprises the development of a quality enhancement technique for image encoders that use compressive sensing. The recommended solution seeks to maximize the perceptual quality based objective function, unlike other sparse representation algorithms that minimizes the error-based objective function. The key idea behind this work is to develop an iterative methodology that works as a modifier for the sparse coefficients. The modification procedure is SSIM-based and has been carried out in an iterative and linear manner. The conducted experiments revealed that the recommended technique works better than another SSIM-based modifier termed the SSIM-inspired OMP (iOMP) in terms of SSIM levels gained. The t-test is also utilized to examine our performance for significance, and the results show that the method works well for any type of image and any size, especially when a data-independent based dictionary is used.

Semi-supervised graph regularized nonnegative matrix factorization with local coordinate for image representation

Nonnegative matrix factorization(NMF) is a powerful image representation algorithm in pattern recognition and data mining. However, the traditional NMF does not utilize any label information, or fail to guarantee the sparse parts-based representation. In this paper, we put forward a semi-supervised local coordinate NMF (SLNMF) algorithm, which incorporate the available label information and the local coordinate constraint into NMF. Particularly, SLNMF makes the learned coefficients sparse by adding the coordinate constraint, and enhance the discriminative ability of different classes by using the label information constraint. Furthermore, in order to extract the geometric structure of the data space, we propose a new semi-supervised graph regularized NMF with local coordinate constraint (SGLNMF) method, which incorporates the graph regularization into SLNMF to enhance the discriminative abilities of data representations. SGLNMF not only reveals the intrinsic geometrical information of the data space, but also takes into account the local coordinate constraint and the label information. Clustering experiments on several standard image datasets demonstrate the effectiveness of our proposed SLNMF and SGLNMF methods compared to the state-of-the-art methods.

Differential evolution based regression algorithm for mathematical representation of electrical parameters in lithium-ion battery model

Accurate equivalent electric circuit models of a lithium-ion battery, generally consisting of lumped electrical parameters, are of utmost importance for implementing various algorithms in Battery Management System (BMS). The low-cost Electronic Control Unit for BMS application prompts for an accurate representation of these parameters with a reduced memory footprint is significant. This prompted the representation of SoC dependant parameters as polynomial, Gaussian, sum of sine and exponential equations through regression algorithms instead of a lookup table. This article proposes Differential Evolution (DE) based Regression algorithm to represent the experimentally pre-identified parameters as a function of SoC through the aforementioned equations. Based on the experimental test, data points representing the SoC dependency of battery parameters such as average open-circuit voltage, hysteresis, internal resistances, and capacitance have been obtained. These data points are subjected to a proposed algorithm to establish the mathematical relationship between the parameters and SoC. The flexibility of the algorithm has been verified by considering the aforesaid equations. The DE based regression algorithm's performance was compared with various other algorithms in terms of statistical parameters such as RMSE, Mean Average, and Variance. Based on the analysis of the obtained results for various algorithms, the equations developed by DE based regression algorithm have the best fit with experimental data for all the aforementioned equations. The analysis depicts that the eighth order polynomial equation fitted by the proposed algorithm is highly accurate. Various battery models have been validated with the experimental data based on the developed equations for all parameters.

Distance Preserving Model Order Reduction of Graph-Laplacians and Cluster Analysis

Graph-Laplacians and their spectral embeddings play an important role in multiple areas of machine learning. This paper is focused on graph-Laplacian dimension reduction for the spectral clustering of data as a primary application, however, it can also be applied in data mining, data manifold learning, etc. Spectral embedding provides a low-dimensional parametrization of the data manifold which makes the subsequent task (e.g., clustering with k-means or any of its approximations) much easier. However, despite reducing the dimensionality of data, the overall computational cost may still be prohibitive for large data sets due to two factors. First, computing the partial eigendecomposition of the graph-Laplacian typically requires a large Krylov subspace. Second, after the spectral embedding is complete, one still has to operate with the same number of data points, which may ruin the efficiency of the approach. For example, clustering of the embedded data is typically performed with various relaxations of k-means which computational cost scales poorly with respect to the size of data set. Also, they become prone to getting stuck in local minima, so their robustness depends on the choice of initial guess. In this work, we switch the focus from the entire data set to a subset of graph vertices (target subset). We develop two novel algorithms for such low-dimensional representation of the original graph that preserves important global distances between the nodes of the target subset. In particular, it allows to ensure that target subset clustering is consistent with the spectral clustering of the full data set if one would perform such. That is achieved by a properly parametrized reduced-order model (ROM) of the graph-Laplacian that approximates accurately the diffusion transfer function of the original graph for inputs and outputs restricted to the target subset. Working with a small target subset reduces greatly the required dimension of Krylov subspace and allows to exploit the conventional algorithms (like approximations of k-means) in the regimes when they are most robust and efficient. This was verified in the numerical clustering experiments with both synthetic and real data. We also note that our ROM approach can be applied in a purely transfer-function-data-driven way, so it becomes the only feasible option for extremely large graphs that are not directly accessible. There are several uses for our algorithms. First, they can be employed on their own for representative subset clustering in cases when handling the full graph is either infeasible or simply not required. Second, they may be used for quality control. Third, as they drastically reduce the problem size, they enable the application of more sophisticated algorithms for the task under consideration (like more powerful approximations of k-means based on semi-definite programming (SDP) instead of the conventional Lloyd’s algorithm). Finally, they can be used as building blocks of a multi-level divide-and-conquer type algorithm to handle the full graph. The latter will be reported in a separate article.