Cyclic Network Research Articles

PathWyse: a flexible, open-source library for the resource constrained shortest path problem

In this paper, we consider a fundamental and hard combinatorial problem: the Resource Constrained Shortest Path Problem (RCSPP). We describe the implementation of a flexible, open-source library for the solution of the RCSPP, called PathWyse, capable of tackling several variants of the problem. We designed PathWyse with the final user in mind, developing easy-to-use interfaces without compromising performance. We provide computational experiments on three classes of instances of the RCSPP, namely RCSPP on cyclic networks, RCSPP on large acyclic networks, and RCSPP on ad-hoc cyclic networks. We show that PathWyse is packed off-the-shelf with algorithms capable of tackling generic problems, and can exploit dedicated algorithms for specific classes. This paper represents the first step along the journey of devising and implementing a comprehensive open-source library for a large variety of RCSPPs. The current version of the library carries exact algorithms for the RCSPP but new algorithms, both heuristic and exact, will be added thanks to the flexible design. We also foresee PathWyse becoming a platform ready for data-driven and process-driven methodologies for these types of problems.

A computational study of H-bonded networks in cyclic water clusters, (H2O)n (n = 3-12).

We have performed a detailed MM and DFT investigation of neutral water clusters (H2O)n (n = 3-12). Our results show the trend of interaction energies in these clusters as a function of the size of the cluster. They show that the H-bond strength increases with cluster size and that the model of water is better described if two different partial charges are used on the hydrogen, depending on whether hydrogen is H-bonded or not. The average binding enthalpy change due to the formation of H-bonds between water molecules is found to be - 25.9kJmol-1 at B3LYP/aug-cc-pVDZ level of theory. We observe the formation of cyclic H-bonded networks through the analysis of frontier orbitals and IR vibrational frequencies spectra. For the water cluster with n = 11, we observe an unusual reduction of the bandgap indicative of a cyclic H-bonded network. Calculations were performed with the MMFF94 force field and the B3LYP method using various large basis sets. Molecular orbital diagrams and population analysis were done using standard tools in Gaussian.

Identification of Intensive Oscillation Parameters of Power System Based on CS-CNN-GRU Network

Broadband oscillations caused by power electronic devices contain a large number of high harmonics and interharmonics, and the frequencies of these harmonic/interharmonic components are likely to be very close. In order to improve the accuracy of dense oscillation parameter identification, this paper proposes a cyclic gate unit-convolutional neural network (CNN-GRU) dense oscillation parameter identification method optimized by the cuckoo search algorithm (CS). The Cuckoo search algorithm combined with the cross-entropy function realizes the automatic acquisition of the parameters of the CNN-GRU network and improves the adaptive ability of the model. The analysis of the measured data shows that the method in this paper has high accuracy in identifying the dense oscillation parameters of dense frequency harmonics, interharmonics, and neighboring fundamental interharmonic pairs, etc., and the computational speed is also improved compared with the traditional deep learning network.

Retracted: Using a Solution Construction Algorithm for Cyclic Shift Network Coding under Multicast Network to the Transformation of Musical Performance Styles

Retracted: Using a Solution Construction Algorithm for Cyclic Shift Network Coding under Multicast Network to the Transformation of Musical Performance Styles

An internet of health things-driven skin cancer classification using progressive cyclical convolutional neural network with ResNexT50 optimized by exponential particle swarm optimization

Skin cancer has become more common in the last few decades according to reports by the World Health Organization. Currently, 1,32,000 melanoma skin cancer cases and around three million non-melanoma cases are reported annually worldwide. Early skin cancer identification with classification allows for effective diagnosis and therapy for patients. Therefore, an internet of health things-driven deep learning approach for identification with classification of skin cancer based on Progressive Cyclical Convolutional Neural Network with ResNexT50 transfer learning optimized by Exponential Particle Swarm Optimization (IoHT-SC-PCCNN-RNT-EXPSO) is proposed in this manuscript. Here, the input skin cancer images are amassed from international Skin Imaging Collaboration (ISIC) image archive dataset. The Dynamic Temporal Filter (DTF) is used for removing the noise, and also enhances the skin cancer images quality. In which, PCCNN does not adopt any optimization methods for determining the optimum parameter to offer accurate skin cancer classification. So, Exponential Particle Swarm Optimization (EXPSO) is proposed for optimizing PCCNN to help it precisely classify the skin cancer. The proposed IoHT-SC-PCCNN-RNT-EXPSO method is implemented in MATLAB. The metrics, like accuracy, recall, precision, f1-Score, specificity, computation time, error rate are evaluated to evaluate the effectiveness of the proposed technique. The proposed MSCC-DI-SDRN-JSOA demonstrated improved specificity with percentages of 27.35 %, 28.64 %, and 24.38 %. Additionally, it showed lower computation time by 37.28 %, 29.39 %, and 28.44 %, and higher accuracy by 30.50 %, 32.39 %, and 31.54 % compared to the existing methods, like IoHT-SC-CNN, IoHT-SC-KNN and IoHT-SC-DWNN-EPO respectively.

Intermolecular hydrogen bond interactions in water clusters of zwitterionic glycine: DFT, MESP, AIM, RDG, and molecular dynamics analysis

The putative global minima geometries of zwitterionic glycine-water complexes [zGly.Wn (1 = 1–10)] are controlled by the cyclic network of water molecules bonded to the zwitterionic glycine. Their stability is governed by the number of water-water interactions along with the existence of the intramolecular glycine-water bonds. The computed intramolecular distances suggest that water molecules tend to coordinate within themselves. The computed association energy attains saturation at 6, implying that up to 6 water molecules are coordinated to the zwitterionic glycine. Further addition of water molecules happens to be an intramolecular addition to the existing water. The computed incremental association energy decreases with the increase in the number of water molecules, except for the six and nine water molecules. This implies that compared to the water clusters, the zwitterionic clusters are less stable. The higher stability of n = 6 and 9 can be marked to the resemblance of the cyclic network of water molecules to the bare global minima water clusters. Benchmarking calculations suggest that M06-2X-D3 functional provides binding energy close to the PW6B97D3(BJ) method. The PGM clusters' thermodynamic parameters show that the reaction's enthalpy remains constant and is negative, indicating that the hydration process is exothermic and independent of the number of water molecules. The electron density difference (EDD) diagram shows the build-up of charge is more pronounced in the intramolecular hydrogen bonds (HBs) between water molecules than between zwitterionic glycine and water. QTAIM analysis shows that the water-water HBs are comparatively weaker than zwitterionic glycine-water interactions.

On the synergetic effects of cyclic cooperativity in water clusters.

The present study delves into the question of how the strength of a hydrogen bond (HB) common to two or more cyclic HB networks is influenced by the cooperativity contributions (CCs) of these cycles. We employ the molecular tailoring approach-based method to calculate the cyclic CCs in water clusters, Wn (n = 6-20). The energy of an HB in a Wn cluster is estimated by adding the total cyclic CC to its counterpart in the respective dimer. The resulting HB energies closely match their full cluster counterparts, typically within 1.0 kcal mol-1, with substantial CCs of these cyclic networks.

Normalized Cyclic Loop Network for Rail Surface Defect Detection Using Knowledge Distillation

Normalized Cyclic Loop Network for Rail Surface Defect Detection Using Knowledge Distillation

Toughening of a polymer network by the addition of a small amount of large-sized multicyclic chains

Topological interactions between cyclic and linear polymers have attracted significant attention to exploring the properties of soft materials. However, the complete picture of the cyclic–linear topological interactions is unknown. In this study, a blend system composed of a polymer network and a small amount of large-sized (multi)cyclic polymers (cyclic chain-mixed network, CMN) was prepared to achieve a reinforcing effect via rotaxane or catenane formation. CMNs can be toughened by tuning the topological variables of the (multi)cyclic chains (size and number of cyclic units) and the cross-link density of the network. The toughening was maximized when the molecular weights of the cyclic units were comparable to those of the network loops and the number of rings per molecule was approximately 5. This article suggests that CMNs can control the mechanical properties of networks containing cyclic chains using a single component and proposes a topological design for fabricating tough CMN materials.

Scroll-growing/controlling chaotic attractors in cyclic Hopfield neural networks via memristive bridging

In this paper, a memristor-based neural network is proposed, which is implemented by two tri-neuron resistive-cyclic Hopfield neural networks (RC-HNNs) via memristive bridging. The memristor-bridged network has a line equilibrium set composed of infinitely many index-2 saddle-foci, but it can produce multi-scroll chaotic attractors contrary to Shil’nikov’s criterion. Complex bifurcation behaviors, scroll-growing chaotic attractors over time, and homogeneous coexisting attractors are revealed by numerical methods. Further, a scroll-control scheme is designed and scroll-controlling chaotic attractors are demonstrated numerically. The results show that the memristor-bridged network can not only generate scroll-growing chaotic attractors over time, but also produce scroll-controlling chaotic attractors by limiting the dynamic range of the internal state of the bridging memristor. Finally, an analog electronic circuit is designed for the memristor-bridged network, and PSIM circuit simulations are used to verify the numerical simulations.

Dissipative Cyclic Reaction Networks: Mechanistic Insights into a Minor Enantiomer Recycling Process

AbstractAn analysis of an out‐of‐equilibrium cyclic reaction network which continuously converts a minor undesired product enantiomer to the desired major enantiomer by irreversible addition of chemical fuel and irreversible elimination of spent fuel is presented. The reaction network is maintained as long as fuel is added; interrupted fuel addition drives the system towards equilibrium, but the cyclic process restarts upon resumed fuel addition, as demonstrated by three consecutive fuel cycles. The process is powered by the hydrolysis of methyl cyanoformate to HCN and monomethyl carbonic acid, which decomposes to CO2 and MeOH. The time it takes to reach steady state depends on the rate of conversion of the fuel and decreases with increased conversion rate. Three catalysts, one metal catalyst and two enzymes, together constitute an efficient regulation system allowing control of the forward, backward and waste‐forming steps, thereby assuring the production of high yields of products with high enantiopurity.

Attribute-Guided Collaborative Learning for Partial Person Re-Identification.

Partial person re-identification (ReID) aims to solve the problem of image spatial misalignment due to occlusions or out-of-views. Despite significant progress through the introduction of additional information, such as human pose landmarks, mask maps, and spatial information, partial person ReID remains challenging due to noisy keypoints and impressionable pedestrian representations. To address these issues, we propose a unified attribute-guided collaborative learning scheme for partial person ReID. Specifically, we introduce an adaptive threshold-guided masked graph convolutional network that can dynamically remove untrustworthy edges to suppress the diffusion of noisy keypoints. Furthermore, we incorporate human attributes and devise a cyclic heterogeneous graph convolutional network to effectively fuse cross-modal pedestrian information through intra- and inter-graph interaction, resulting in robust pedestrian representations. Finally, to enhance keypoint representation learning, we design a novel part-based similarity constraint based on the axisymmetric characteristic of the human body. Extensive experiments on multiple public datasets have shown that our model achieves superior performance compared to other state-of-the-art baselines.

Cyclic style generative adversarial network for near infrared and visible light face recognition

Near Infrared and Visible Light (NIR-VIS) face recognition attracts attention from researchers because of its potential for safety, illumination invariance, and stability. Nevertheless, the difference between NIR and VIS domains, the domain gap, remains a huge problem for matching NIR and VIS images. Specifically, for the same identity, the appearance in the NIR domain is different from the VIS domain. Thus, traditional face recognition methods cannot be effective. To address this problem, this paper proposes a novel model, called Cyclic-Style GAN (CS-GAN). First, the pre-trained Style-GAN 3 network is embedded into the Cycle-GAN structure for NIR-VIS cross-domain learning. Second, there is a cyclic subspace learning method consisting of latent loss and style loss, through which both style (domain feature) and facial characteristic features are learned to improve the quality of synthesized images. The model synthesizes realistic VIS images from NIR images and does the face recognition task in the VIS domain. The proposed method achieves 99.6% Rank-1 accuracy on the CASIA NIR-VIS 2.0 database which is a state-of-the-art result. The visualization results show that the proposed model synthesizes VIS images with a clear texture of faces and in close-to-reality color.

Soybean Leaf Diseases Recognition Based on Generative Adversarial Network and Transfer Learning

Soybean leaf disease labeling data are not easy to obtain, and soybean leaf disease model training often needs a lot of data. Due to the limitations of fixed rules such as rotation and clipping, traditional data enhancement cannot generate images with diversity and variability. In view of the above problems, this study proposed a data enhancement method based on generative adversarial network to expand the original soybean leaf disease dataset. This method was based on cyclic confrontation network, and its discriminator uses dense connection strategy to realize feature reuse, so as to reduce the amount of calculation. In the training process, improved transfer learning is used to automatically fine tune the pre-training model. The accuracy of the optimized method in 9 kinds of soybean leaf disease image recognition is 95.84%, which is 0.98% higher than the traditional fine-tuning method. The experimental results show that this method based on generating confrontation network has significant ability in generating soybean leaf disease image, and can expand the existing dataset. In addition, this method also provides an effective data enhancement solution for the expansion of other crop disease image datasets.

Endogenous production networks with fixed costs

We develop a model of endogenous production networks with fixed costs in the formation of links between firms. We show that the closed economy equilibrium is unique if the set of feasible networks consists only of networks that are acyclic and the buyer initiates the link formation while having full bargaining power in price negotiations with the supplier. We provide examples of multiple equilibria if the supplier initiates the link formation in both cyclic and acyclic feasible networks or if the buyer initiates the link formation in a cyclic production network. We take the acyclic production network model to Belgian data on firm-to-firm production networks and show that it matches well the salient features of the network. The model generates substantial churn in domestic firm-to-firm linkages in response to trade shocks, while delivering only moderately different welfare changes compared to a model with fixed linkages.

DRGAN: Dense Residual Generative Adversarial Network for Image Enhancement in an Underwater Autonomous Driving Device.

Underwater autonomous driving devices, such as autonomous underwater vehicles (AUVs), rely on visual sensors, but visual images tend to produce color aberrations and a high turbidity due to the scattering and absorption of underwater light. To address these issues, we propose the Dense Residual Generative Adversarial Network (DRGAN) for underwater image enhancement. Firstly, we adopt a multi-scale feature extraction module to obtain a range of information and increase the receptive field. Secondly, a dense residual block is proposed, to realize the interaction of image features and ensure stable connections in the feature information. Multiple dense residual modules are connected from beginning to end to form a cyclic dense residual network, producing a clear image. Finally, the stability of the network is improved via adjustment to the training with multiple loss functions. Experiments were conducted using the RUIE and Underwater ImageNet datasets. The experimental results show that our proposed DRGAN can remove high turbidity from underwater images and achieve color equalization better than other methods.

CPSNet: a cyclic pyramid-based small lesion detection network

CPSNet: a cyclic pyramid-based small lesion detection network

Single-Photon Ionization Induced New Covalent Bond Formation in Acrylonitrile(AN)-Pyrrole(Py) Clusters.

The formation of nitrogen-containing organic compounds is crucial for understanding chemical evolution and the origin of life in the interstellar medium (ISM). In this study, we explore whether acrylonitrile (AN) and pyrrole (Py) can form new nitrogen-containing compounds after single-photon ionization in their gaseous clusters by vacuum ultraviolet (VUV)-infrared (IR) spectroscopy and theoretical calculations. The results show that a strong linear H-bond is formed in neutral AN-Py, while cyclic or bicyclic H-bonded networks are formed in the neutral AN-Py2 cluster. It is found that the structure containing a new C-C covalent bond between two moieties in (AN-Py)+ is formed besides the formation of H-bonded structures after AN-Py is ionized by VUV light. In (AN-Py2)+ cluster cations, new C-C or C-N covalent bonds tend to be formed between two Py, with (Py)2+ as the core in the cluster. The results reveal that new covalent bonds are more likely to be formed between two Py species when AN and Py are present in the cationic clusters. These results provide spectroscopic evidence of the formation of new nitrogen-containing organic compounds from AN and Py induced by VUV, which are helpful for our understanding of the formation of diverse prebiotic molecules in interstellar space.

Word Embedding for Text Classification: Efficient CNN and Bi-GRU Fusion Multi Attention Mechanism

The proposed methodology for the task of text classification involves the utilization of a deep learning algorithm that integrates the characteristics of a fusion model. The present model is comprised of several attention-based Convolutional Neural Networks (CNNs) and Gate Recurrent Units (GRUs) that are organized in a cyclic neural network. The Efficient CNN and Bi-GRU Fusion Multi Attention Mechanism is a method that integrates convolutional neural networks (CNNs) and bidirectional Gated Recurrent Units (Bi-GRUs) with multi-attention mechanisms in order to enhance the efficacy of word embedding for the purpose of text classification. The proposed design facilitates the extraction of both local and global features of textual feature words and employs an attention mechanism to compute the significance of words in text classification. The fusion model endeavors to enhance the performance of text classification tasks by effectively representing text documents through the combination of CNNs, Bi-GRUs, and multi-attention mechanisms. This approach aims to capture both local and global contextual information, thereby improving the model’s ability to process and analyze textual data. Moreover, the amalgamation of diverse models can potentially augment the precision of text categorization. The study involved conducting experiments on various data sets, including the IMDB film review data set and the THUCNews data set. The results of the study demonstrate that the proposed model exhibits superior performance compared to previous models that relied solely on CNN, LSTM, or fusion models that integrated these architectures. This superiority is evident in terms of accuracy, recall rate, and F1 score.

A Novel Event-Driven Spiking Convolutional Neural Network for Electromyography Pattern Recognition.

Electromyography (EMG) pattern recognition is an important technology for prosthesis control and human-computer interaction etc. However, the practical application of EMG pattern recognition is hampered by poor accuracy and robustness due to electrode shift caused by repeated wearing of the signal acquisition device. Moreover, the user's acceptability is low due to the heavy training burden, which is caused by the need for a large amount of training data by traditional methods. In order to explore the advantage of spiking neural network (SNN) in solving the poor robustness and heavy training burden problems in EMG pattern recognition, a spiking convolutional neural network (SCNN) composed of cyclic convolutional neural network (CNN) and fully connected modules is proposed and implemented in this study. High density surface electromyography (HD-sEMG) signals collected from 6 gestures of 10 subjects at 6 electrode positions are taken as the research object. Compared to CNN with the same structure, CNN-Long Short Term Memory (CNN-LSTM), linear kernel linear discriminant analysis classifier (LDA) and spiking multilayer perceptron (Spiking MLP), the accuracy of SCNN is 50.69%, 33.92%, 32.94% and 9.41% higher in the small sample training experiment, 6.50%, 4.23%, 28.73%, and 2.57% higher in the electrode shifts experiment respectively. In addition, the power consumption of SCNN is about 1/93 of CNN. The advantages of the proposed framework in alleviating user training burden, mitigating the adverse effect of electrode shifts and reducing power consumption make it very meaningful for promoting the development of user-friendly real-time myoelectric control system.