Effective Representation Research Articles

HybAVPnet: A Novel Hybrid Network Architecture for Antiviral Peptides Prediction.

Viruses pose a great threat to human production and life, thus the research and development of antiviral drugs is urgently needed. Antiviral peptides play an important role in drug design and development. Compared with the time-consuming and laborious wet chemical experiment methods, it is critical to use computational methods to predict antiviral peptides accurately and rapidly. However, due to limited data, accurate prediction of antiviral peptides is still challenging and extracting effective feature representations from sequences is crucial for creating accurate models. This study introduces a novel two-step approach, named HybAVPnet, to predict antiviral peptides with a hybrid network architecture based on neural networks and traditional machine learning methods. We adopted a stacking-like structure to capture both the long-term dependencies and local evolution information to achieve a comprehensive and diverse prediction using the predicted labels and probabilities. Using an ensemble technique with the different kinds of features can reduce the variance without increasing the bias. The experimental result shows HybAVPnet can achieve better and more robust performance compared with the state-of-the-art methods, which makes it useful for the research and development of antiviral drugs. Meanwhile, it can also be extended to other peptide recognition problems because of its generalization ability.

The incorporation of stacked long short-term memory into intrusion detection systems for botnet attack classification

<p><span lang="EN-US">Botnets are a common cyber-attack method on the internet, causing infrastructure damage, data theft, and malware distribution. The continuous evolution and adaptation to enhanced defense tactics make botnets a strong and difficult threat to combat. In light of this, the study's main objective was to find out how well techniques like principal component analysis (PCA), synthetic minority oversampling technique (SMOTE), and long short-term memory (LSTM) can help find botnet attacks. PCA shows the ability to reduce the feature dimensions in network data, allowing for a more efficient and effective representation of the patterns contained. The SMOTE addresses class imbalances in the dataset, enhancing the model's ability to recognize suspicious activity. Furthermore, LSTM classifies sequential data, understanding complex network patterns and behaviors often used by botnets. The combination of these three methods provided a substantial improvement in detecting suspicious botnet activities. We also evaluated the effectiveness using performance metrics such as accuracy, precision, recall, and F1-score. The results showed an accuracy of 96.77%, precision of 88.95%, recall of 88.58%, and F1-score of 88.64%, indicating that the proposed model was reliable in detecting botnet traffic compared to other deep learning models. Furthermore, LSTM can classify sequential data, understanding complex network patterns and behaviors often used by botnets.</span></p>

Constraints on fracture connectivity from amplitude variations with incidence angle and azimuth analysis

The analysis of seismic reflection amplitude variations with incidence angle and azimuth (AVAz) is a powerful tool for studying and characterizing fractured formations. There is evidence to suggest that wave-induced fluid pressure diffusion (FPD) within connected fractures can significantly affect the AVAz response. This, in turn, indicates that this seismic attribute may contain information on the connectivity degree of the fracture network, which represents a key hydraulic property of fractured formations. The assessment of the sensitivity of seismic reflectivity with respect to fracture connectivity accounting for FPD effects to date is limited to 2D models, which is mainly due to the complexity and the associated computational cost of modeling the effective properties of the corresponding generically anisotropic 3D media. To overcome these issues, we use a numerical upscaling procedure based on Biot’s theory of poroelasticity that allows us to obtain effective anisotropic representations of 3D fractured formations accounting for the FPD effects. Using these effective properties, we then perform a plane-wave analysis to compute the reflection coefficients at the top of the corresponding fractured formation. The results of our numerical analyses extend the previous 2D studies and demonstrate that seismic reflection coefficients are highly sensitive to the degree of connectivity of the fracture networks. The reflectivity responses of formations comprising connected and unconnected fractures indicate pronounced differences not only with respect to incidence angle but also with respect to azimuth. Quite importantly, our results also demonstrate that crossplots of commonly used AVAz coefficients permit us to identify regions with higher or lower fracture connectivity.

Reconstruction of large-scale anisotropic 3D digital rocks from 2D shale images using generative adversarial network

Digital rock technology has played a significant role in unconventional oil and gas exploration and development. Current techniques face challenges such as high cost or low efficiency when reconstructing 3D pore structures at the nanoscale, which limit the study of rock physical properties in reservoirs. To address these issues, this work proposes an efficient generative adversarial network workflow for generating large-scale anisotropic 3D digital rocks from 2D images. Instead of relying on 3D data, the model is trained using only 2D images. During training, rock images taken from various angles can be utilized to construct anisotropic 3D digital rocks. Moreover, a novel method for generating large-scale 3D data is devised, where the input random noise is divided with overlap to generate small-scale digital rocks and concatenated into a larger-scale 3D digital rock. The similarity between the 3D reconstructed data and the training images was assessed separately, as well as the pore structure of the 3D reconstructed data, and the feasibility of the proposed method was verified using the simulation of absolute permeability. The digital rock modeling process presented in this study enables rapid and accurate 2D-to-3D large-scale digital rock modeling, providing an effective representation of the anisotropy present in shale oil reservoirs.

Siamese Tracking Network with Multi-attention Mechanism

Object trackers based on Siamese networks view tracking as a similarity-matching process. However, the correlation operation operates as a local linear matching process, limiting the tracker’s ability to capture the intricate nonlinear relationship between the template and search region branches. Moreover, most trackers don’t update the template and often use the first frame of an image as the initial template, which will easily lead to poor tracking performance of the algorithm when facing instances of deformation, scale variation, and occlusion of the tracking target. To this end, we propose a Simases tracking network with a multi-attention mechanism, including a template branch and a search branch. To adapt to changes in target appearance, we integrate dynamic templates and multi-attention mechanisms in the template branch to obtain more effective feature representation by fusing the features of initial templates and dynamic templates. To enhance the robustness of the tracking model, we utilize a multi-attention mechanism in the search branch that shares weights with the template branch to obtain multi-scale feature representation by fusing search region features at different scales. In addition, we design a lightweight and simple feature fusion mechanism, in which the Transformer encoder structure is utilized to fuse the information of the template area and search area, and the dynamic template is updated online based on confidence. Experimental results on publicly tracking datasets show that the proposed method achieves competitive results compared to several state-of-the-art trackers.

[formula omitted] electronic structure and spin EPR-shift of [formula omitted] ion in diluted magnetic p-type [formula omitted][formula omitted]

In this correspondence, we deduce the non-parabolic band dispersion within the framework of effective mass representation (EMR) using k→⋅π→ perturbation theory in the presence of s.o interaction, magnetic impurity and external magnetic field. We measure the electronic parameters like Fermi energy, density of states, effective g factor and effective mass etc. in p-type Sn1−xEuxTe by taking into account a 6-level band structure with normal and bumped band edges. In the study, we use the experimental and simulated band gap values as functions of Eu+2 impurity and temperature. Apart from the general trend, the electronic parameters exhibit extreme behaviour near the band inversion point x=0.020 at T=300 K (Nayak et. al.2023), but in the range of impurity concentration and temperature considered; from x=0.0005 to x=0.001 and T=1.5 K to T=300 K, the behaviour is normal. Again, we follow the expression for spin-electron paramagnetic resonance (EPR) shift (Ps) where the perturbation expansion of Green’s function is used in the presence of the above-mentioned interactions. To get around the problem of loss of translational symmetry in the presence of the magnetic field, Green’s function is redefined in terms of Peierl’s phase factor. Here we have made an extensive study of the shift of Eu+2 ion in p-type Sn1−xEuxTe for several impurity concentrations from x=0.0005 to x=0.001 in the temperature range T=1.4 K to T=300 K. Lastly, we analyze the EPR shift and other electronic parameters and their anisotropies in p-type Sn1−xEuxTe as functions of carrier concentration and Eu impurity from T=1.5K to T=300K. We report the EPR shift is 2.97×10−3 at T=300 K and 3.31×10−3 at T=1.5 K with anisotropy at p=5×1020cm−3 in contrast to the experimental observation of +(3.4±0.4)×10−3 in the same system. The discrepancy is attributed to the reduction in the density of available states at Fermi energy at the L-band caused by the simultaneous rise in that at the Σ− band noticeable particularly at/above high hole concentrations p≥5×1020cm−3.

BCLNet: Boundary contrastive learning with gated attention feature fusion and multi-branch spatial-channel reconstruction for land use classification

The fusion of optical and synthetic aperture radar (SAR) images is not only a crucial method for enhancing land use classification tasks but also forms a fundamental basis for the interpretation of multimodal remote sensing imagery. However, the visual differences caused by the distinct nonlinear radiometric properties of optical and SAR images lead to challenges in effectively integrating multimodal features using current land use classification methods, resulting in lower classification accuracy. On one hand, these approaches may be affected by the heterogeneity among different source images, leading to an inability to fully utilize the semantic content of the fused features. On the other hand, the semantic embedding space does not consider the semantic relationships between different pixels from a global perspective (i.e., across the entire dataset). To address the aforementioned challenges, we propose the boundary contrastive learning classification network (BCLNet). In this framework, we introduce the gate attention fusion (GAF) module, which selectively considers distinctiveness and consistency features in optical and SAR imagery, facilitating effective modal representation fusion. Furthermore, we introduce the multi-branch spatial channel reconstruction (MSCR) module for enhanced feature augmentation, operating in both spatial and channel dimensions with a dual-dimensional feature reconstruction selection mechanism for the fused features. Furthermore, we adopt boundary contrast learning to address the global pixel semantic space embedding problem. Our method is evaluated on the WHU-OPT-SAR dataset and a multi-class scene classification dataset (MSD) that we constructed. Compared to state-of-the-art methods, our approach demonstrates improvements in pixel classification overall accuracy (OA) and mean intersection over union (mIOU).

Transformer-based medication recommendation with a multiple graph augmentation strategy

Medication recommendation (MR) is a promising task that benefits both patients and medical practitioners. However, the individual differences in patients and drug–drug interactions (DDIs) in medication combinations are two key factors that severely limit the personalization and safety of the present MR models. To alleviate the challenges mentioned above, this paper proposes a novel MR model named Trans-GAHNet, which innovatively leverages transformer-based representation learning and constructs multiple graphs for drug information augmentation. More specifically, the transformer network architecture is deployed to fully encode the sequential EHR data for effective patient representations, which addresses the cold start issue and enhances the interpretability of representation learning. Next, multiple graphs are employed to capture the co-occurrence and exclusiveness in drug pairs and to mitigate potential DDIs in drug combinations, thereby enhancing the personalization and safety of the predicted medication combinations. Finally, extensive experiments on real-world datasets are conducted to evaluate the proposed model, and the experimental results demonstrate that Trans-GAHNet outperforms the state-of-the-art baseline models on multiple metrics, with Jaccard, F1 and PRAUC scores of 0.5238, 0.6786 and 0.7795, respectively.

Math for Everybody: A Sonification Module for Computer Algebra Systems Aimed at Visually Impaired People

Computer Algebra Systems (CAS) currently lack an effective auditory representation, with most existing solutions relying on screen readers that provide limited functionality. This limitation prevents blind users from fully understanding and interpreting mathematical expressions, leading to confusion and self-doubt. This paper addresses the challenges blind individuals face when comprehending mathematical expressions within a CAS environment. We propose “Math for Everybody” (Math4e, version 1.0), a software module to reduce barriers for blind users in education. Math4e is a Sonification Module for CAS that generates a series of auditory tones, prosodic cues, and variations in audio parameters such as volume and speed. These resources are designed to eliminate ambiguity and facilitate the interpretation and understanding of mathematical expressions for blind users. To assess the effectiveness of Math4e, we conducted standardized tests employing the methodologies outlined in the Software Engineering Body of Knowledge (SWEBOK), International Software Testing Qualifications Board (ISTBQ), and ISO/IEC/IEEE 29119. The evaluation encompassed two scenarios: one involving simulated blind users and another with real blind users associated with the “Asociación de Invidentes Milton Vedado” foundation in Ecuador. Through the SAM methodology and verbal surveys (given the condition of the evaluated user), results are obtained, such as 90.56% for pleasure, 90.78% for arousal, and 91.56% for dominance, which demonstrates significant acceptance of the systems by the users. The outcomes underscored the users’ commendable ability to identify mathematical expressions accurately.

SMDE: Unsupervised representation learning for time series based on signal mode decomposition and ensemble

Time series data pervades multiple domains, yet in real-world applications, there frequently exists a deficiency of labels to ascertain effective representations and enable efficient classification. In this work, we introduce SMDE, a unique time series contrastive learning framework, diverging from conventional techniques that predominantly concentrate on instance-level contrast. SMDE leverages signal decomposition and ensemble methodologies, enhancing our focus on localized mode alterations, within time series signals. We further introduce the global signal consistency and intrinsic mode consistency mechanisms to promote the robust learning of global and local attributes. To ensure more efficient convergence, we present the DE Circle loss furnished with an adaptive weighting mechanism, founded on global signal consistency and intrinsic mode consistency principles. The SMDE framework is designed to be applicable to both univariate and multivariate time series analysis, indicating its potential for a wide range of temporal data applications. Experiments on 128 univariate UCR datasets reveal that SMDE improves the average accuracy by 4.73% over existing techniques. Additionally, our experiments on 29 multivariate UEA datasets demonstrate an average accuracy improvement of 7.42% compared to current methodologies. Moreover, through extensive semi-supervised experimentation, SMDE demonstrates superior performance in 15 disparate semi-supervised configurations, attaining an equivalent performance level to supervised training with 100% labeled data, even with a minimal portion of the labels. The code is publicly available at https://github.com/haobinlaosi/smde.

Combined effects of fine and coarse marine aerosol on vertical raindrop size distribution

Climate models commonly overestimate warm rain frequency and underestimate its intensity over the ocean, primarily due to insufficient representation of the aerosol effects. This pertains to both fine aerosols (FA) and coarse sea spray aerosols (CSA), where the latter is mostly absent in the models. Here, our observations show that adding CSA enhances vertical warm rain structure, in contrast to the effect of FA. The magnitude of the effect of CSA is larger than the opposite effect of the FA. For rain with top heights of 2–3 km, the raindrop size, concentration, and rain rate can be increased by factors of 1.03, 1.47, and 1.60, respectively. These CSA-induced changes are larger for thicker clouds, reaching a maximum by factors of 1.12, 1.85, and 2.21, respectively. Therefore, the combined FA and CSA effects should be incorporated into climate models for accurately simulated precipitation microphysical processes.

Design Optimization of a THz Receiver Based on 60 nm Complementary Metal–Oxide–Semiconductor Technology

The technology transfer of terahertz wireless communication from research laboratories to commercial applications is a global strategic achievement currently pursued to match the ever-increasing demand for high-speed communication. The use of commercial integrated electronics for the detection of THz waves is an intriguing challenge which has enticed great interest in the scientific research community. Rapid progress in this field has led to the exploitation of THz direct detection using standard CMOS technology based on the so-called self-mixing effect. Our research, stemming out of a collaboration between Sapienza University of Rome and STMicroelectronics company, is focused on the complete design process of a THz rectifier, realized using 50 nm ST B55 CMOS technology. In this paper, we report the optimization process of a case-study receiver, aimed to demonstrate the feasibility of direct demodulation of the transmitted OOK signal. A relatively limited bandwidth extension is considered since the device will be included in a system adopting a radiation source with a limited band. The design refers to a specific technology, the 60 nm MOS in B55X ST; nevertheless, the proposed optimization procedure can be applied in principle to any MOS device. Several aspects of the rectification process and of the receiver design are investigated by combining different numerical simulation methodologies. The direct representation of the rectification effect through the equivalent circuit of the detector is provided, which allows for the investigation of the detector–amplifier coupling, and the computation of output noise equivalent power. Numerical results are presented and used as the basis for the optimization of the receiver parameters.

Sustainable Transportation: Exploring the Node Importance Evolution of Rail Transit Networks during Peak Hours

The scientific and rational assessment of the evolution of node importance in rail transit line networks is important for the sustainability of transportation systems. Based on the complex network theory, this study develops a weighted network model using the Space L method. It first considers the network topology, the mutual influence of neighboring nodes of the transportation system, and the land use intensity in the station influence domain to construct a comprehensive index evaluation system of node importance. It then uses the covariance-weighted principal component analysis algorithm to more comprehensively evaluate the node importance evolution mechanism and analyzes the similarity and difference of the sorting set by adopting three different methods. The interaction mechanism between the distribution of important nodes and the evolution of land use intensity is explored in detail based on the fractal dimension theory. The Xi’an rail transit network is considered an example of qualitative and quantitative analysis. The obtained results show that the importance of nodes varies at different times of the day and the complexity of the morning peak is more prominent. Over time, articulated fragments with significance values greater than 0.5 are formed around the station, which are aligned with the direction of urban development, creating a sustainable mechanism of interaction. As the network’s crucial nodes in the center of gravity increase and the southern network expands, along with the increased intensity of the city’s land utilization, the degree of alignment in evolution becomes increasingly substantial. Different strategies for attaching the network, organized based on the size of Si can lead to the rapid damage of the network (reducing it to 0.2). The identification of crucial nodes highlighted in this paper serves as an effective representation of the functional characteristics of the nodes in transportation networks. The results obtained can provide a reference for the operation and management of metro systems and further promote the sustainable development of transportation networks.

TSEDNet:Task-specific encoder–decoder network for surface defects of strip steel

Deep learning faces challenges in the surface defect segmentation of strip steel. Firstly, insufficient processing of feature maps leads to the loss of task-specific feature information. Secondly, the segmentation of defects with long-tail distributions is not accurate enough. To address these issues, a pixel-level deep segmentation method called task-specific encoder–decoder network (TSEDNet) is proposed to construct an end-to-end defect segmentation model. TSEDNet includes the encoder-multi-decoder structure based on domain knowledge settings tailored to specific tasks, which can achieve effective feature representation and significantly reduce the impact of imbalanced defect quantities. Additionally, a novel metric learning method is introduced to optimize decoder selection. Furthermore, the feature fusion module based on metric learning is proposed to utilize general features for restoring task-specific details, thereby enhancing pixel-level segmentation accuracy. Through experiments and industrial validation, the defect segmentation network demonstrates superior performance compared to other advanced segmentation methods and proves its applicability in practical scenarios.

Robust optimization approaches in inventory management: Part B - the comparative study

This comparative study, constituting Part B of our extensive investigation into Robust Optimization (RO) in inventory management, builds on the foundational insights from Part A’s survey. It conducts a thorough analysis of various RO formulations and algorithms, emphasizing their practical application, efficacy, and computational considerations in inventory management contexts. This study is meticulously structured to address the pivotal inquiries identified in Part A, encompassing the effective representation of uncertainty, the selection of optimal decision criteria, the influence of decision rules on inventory management performance, computational challenges, and the adaptability of these methods to evolving technological and market conditions. This research juxtaposes theoretical findings with empirical data, offering a comprehensive evaluation of the strengths, limitations, and practical implications of each robust inventory model discussed in the literature. This investigation not only complements the survey in Part A but also serves as a standalone, in-depth contribution to robust inventory management.

Enhancing clustering representations with positive proximity and cluster dispersion learning

Contemporary deep clustering approaches often rely on contrastive or non-contrastive techniques to acquire effective representations for clustering tasks. Contrastive methods apply negative pairs to achieve homogenous representations but can introduce class collision problems, potentially compromising clustering performance. In contrast, non-contrastive techniques prevent class collisions but may produce non-uniform representations that lead to clustering collapse. In this work, we propose a novel end-to-end deep clustering approach named PIPCDR, designed to harness the strengths of both approaches while mitigating their limitations. PIPCDR incorporates a positive instance proximity loss and a cluster dispersion regularizer. The positive instance proximity loss ensures alignment between augmented views of instances and their sampled neighbors, enhancing within-cluster compactness by selecting genuinely positive pairs within the embedding space. Meanwhile, the cluster dispersion regularizer maximizes inter-cluster distances while minimizing within-cluster compactness, promoting uniformity in the learned representations. PIPCDR excels in producing well-separated clusters, generating uniform representations, avoiding class collision issues, and enhancing within-cluster compactness. We extensively validate the effectiveness of PIPCDR within an end-to-end Majorize-Minimization framework, demonstrating its competitive performance on moderate-scale clustering benchmark datasets and establishing new state-of-the-art results on large-scale datasets.

Attribute Feature Perturbation-Based Augmentation of SAR Target Data.

Large-scale, diverse, and high-quality data are the basis and key to achieving a good generalization of target detection and recognition algorithms based on deep learning. However, the existing methods for the intelligent augmentation of synthetic aperture radar (SAR) images are confronted with several issues, including training instability, inferior image quality, lack of physical interpretability, etc. To solve the above problems, this paper proposes a feature-level SAR target-data augmentation method. First, an enhanced capsule neural network (CapsNet) is proposed and employed for feature extraction, decoupling the attribute information of input data. Moreover, an attention mechanism-based attribute decoupling framework is used, which is beneficial for achieving a more effective representation of features. After that, the decoupled attribute feature, including amplitude, elevation angle, azimuth angle, and shape, can be perturbed to increase the diversity of features. On this basis, the augmentation of SAR target images is realized by reconstructing the perturbed features. In contrast to the augmentation methods using random noise as input, the proposed method realizes the mapping from the input of known distribution to the change in unknown distribution. This mapping method reduces the correlation distance between the input signal and the augmented data, therefore diminishing the demand for training data. In addition, we combine pixel loss and perceptual loss in the reconstruction process, which improves the quality of the augmented SAR data. The evaluation of the real and augmented images is conducted using four assessment metrics. The images generated by this method achieve a peak signal-to-noise ratio (PSNR) of 21.6845, radiometric resolution (RL) of 3.7114, and dynamic range (DR) of 24.0654. The experimental results demonstrate the superior performance of the proposed method.

ICESat-2 Shows Sea Ice Leads Have Little Overall Effects on the Arctic Cloudiness in Cold Months

Abstract The effect of leads in Arctic sea ice on clouds is a potentially important climate feedback. We use observations of clouds and leads from the Ice, Cloud and Land Elevation Satellite-2 (ICESat-2) to study the effects of leads on clouds. Both leads and clouds are strongly forced by synoptic weather conditions, with more clouds over both leads and sea ice at lower sea level pressure. Contrary to previous studies, we find that the overall lead effect on low-level cloud cover is −0.02, a weak cloud dissipating effect in cold months, after the synoptic forcing influence is removed. This is due to compensating contributions from the cloud dissipating effect by newly frozen leads under high pressure systems and the cloud enhancing effect by newly open leads under low pressure system. The lack of proper representation of lead effect on clouds in current climate models and reanalyses may impact their performance in winter months, such as in sea ice growth and Arctic cyclone development.

GraphCL-DTA: A Graph Contrastive Learning With Molecular Semantics for Drug-Target Binding Affinity Prediction.

Drug-target binding affinity prediction plays an important role in the early stages of drug discovery, which can infer the strength of interactions between new drugs and new targets. However, the performance of previous computational models is limited by the following drawbacks. The learning of drug representation relies only on supervised data without considering the information in the molecular graph itself. Moreover, most previous studies tended to design complicated representation learning modules, while uniformity used to measure representation quality is ignored. In this study, we propose GraphCL-DTA, a graph contrastive learning with molecular semantics for drug-target binding affinity prediction. This graph contrastive learning framework replaces the dropout-based data augmentation strategy by performing data augmentation in the embedding space, thereby better preserving the semantic information of the molecular graph. A more essential and effective drug representation can be learned through this graph contrastive framework without additional supervised data. Next, we design a new loss function that can be directly used to adjust the uniformity of drug and target representations. By directly optimizing the uniformity of representations, the representation quality of drugs and targets can be improved. The effectiveness of the above innovative elements is verified on two real datasets, KIBA and Davis. Compared with the GraphDTA model, the relative improvement of the GraphCL-DTA model on the two datasets is 2.7% and 4.5%. The graph contrastive learning framework and uniformity function in the GraphCL-DTA model can be embedded into other computational models as independent modules to improve their generalization capability.

Self-Attention-Assisted TinyML With Effective Representation for UWB NLOS Identification

Self-Attention-Assisted TinyML With Effective Representation for UWB NLOS Identification