Propagation Network Research Articles

Stable feature selection utilizing Graph Convolutional Neural Network and Layer-wise Relevance Propagation for biomarker discovery in breast cancer

High-throughput technologies are becoming increasingly important in discovering prognostic biomarkers and in identifying novel drug targets. With Mammaprint, Oncotype DX, and many other prognostic molecular signatures breast cancer is one of the paradigmatic examples of the utility of high-throughput data to deliver prognostic biomarkers, that can be represented in a form of a rather short gene list. Such gene lists can be obtained as a set of features (genes) that are important for the decisions of a Machine Learning (ML) method applied to high-dimensional gene expression data. Several studies have identified predictive gene lists for patient prognosis in breast cancer, but these lists are unstable and have only a few genes in common. Instability of feature selection impedes biological interpretability: genes that are relevant for cancer pathology should be members of any predictive gene list obtained for the same clinical type of patients. Stability and interpretability of selected features can be improved by including information on molecular networks in ML methods. Graph Convolutional Neural Network (GCNN) is a contemporary deep learning approach applicable to gene expression data structured by a prior knowledge molecular network. Layer-wise Relevance Propagation (LRP) and SHapley Additive exPlanations (SHAP) are methods to explain individual decisions of deep learning models. We used both GCNN+LRP and GCNN+SHAP techniques to construct feature sets by aggregating individual explanations. We suggest a methodology to systematically and quantitatively analyze the stability, the impact on the classification performance, and the interpretability of the selected feature sets. We used this methodology to compare GCNN+LRP to GCNN+SHAP and to more classical ML-based feature selection approaches. Utilizing a large breast cancer gene expression dataset we show that, while feature selection with SHAP is useful in applications where selected features have to be impactful for classification performance, among all studied methods GCNN+LRP delivers the most stable (reproducible) and interpretable gene lists.

Discriminative Action Snippet Propagation Network for Weakly Supervised Temporal Action Localization

Weakly supervised temporal action localization (WTAL) aims to classify and localize actions in untrimmed videos with only video-level labels. Recent studies have attempted to obtain more accurate temporal boundaries by exploiting latent action instances in ambiguous snippets or propagating representative action features. However, empirically handcrafted ambiguous snippet extraction and the imprecise alignment of representative snippet propagation lead to challenges in modeling the completeness of actions for these methods. In this article, we propose a Discriminative Action Snippet Propagation Network (DASP-Net) to accurately discover ambiguous snippets in videos and propagate discriminative instance-level features throughout the video for improving action completeness. Specifically, we introduce a novel discriminative feature propagation module for capturing the global contextual attention and propagating the action concept across the whole video by perceiving the discriminative action snippets with instance information from the same video. Simultaneously, we incorporate denoised pseudo-labels as supervision, where we correct the controversial prediction based on the feature space distribution during training, thereby alleviating false detection caused by noise background features. Furthermore, we design an ambiguous feature mining module, which maximizes the feature affinity information of action and background in ambiguous snippets to generate more accurate latent action and background snippets and learns more precise action instance boundaries through contrastive learning of action and background snippets. Extensive experiments show that DASP-Net achieves state-of-the-art results on THUMOS14 and ActivityNet1.2 datasets.

A network-based method for associating genes with autism spectrum disorder.

Autism spectrum disorder (ASD) is a highly heritable complex disease that affects 1% of the population, yet its underlying molecular mechanisms are largely unknown. Here we study the problem of predicting causal genes for ASD by combining genome-scale data with a network propagation approach. We construct a predictor that integrates multiple omic data sets that assess genomic, transcriptomic, proteomic, and phosphoproteomic associations with ASD. In cross validation our predictor yields mean area under the ROC curve of 0.87 and area under the precision-recall curve of 0.89. We further show that it outperforms previous gene-level predictors of autism association. Finally, we show that we can use the model to predict genes associated with Schizophrenia which is known to share genetic components with ASD.

Forecasting and alert of atmospheric bioaerosol concentration profile based on adaptive genetic algorithm back propagation neural network, atmospheric parameter and fluorescence lidar

Bioaerosols are biologically originated particles in the atmosphere, which is mainly composed of bacteria, fungi, viruses, pollen, spores, and the fragmentation and disintegration of plants and animals. Bioaerosols are easy to be spread in the lower atmosphere and cause various epidemic diseases, which is harmful to human health. The forecasting and alert of bioaerosols have important scientific significance and reality needs. In this paper, a method is proposed for estimating and predicting the concentration profile of atmospheric bioaerosols using fluorescence lidar observational data. Using the powerful nonlinear prediction ability of artificial neural networks and through repeated training, a mathematical model can be established for the relationship among atmospheric environment, meteorological parameters, and bioaerosol concentration profiles. The input parameters are temperature and humidity, aerosol extinction coefficient, backscatter coefficient, PM2.5, PM10, SO2, NO2, CO, O3, and wind speed, and outputs the concentration profile of bioaerosols. The prediction results with the measurement relative deviation of genetic algorithm back propagation (GA-BP) neural network and adaptive genetic algorithm back propagation (AGA-BP) neural network were analyzed. The results indicate that the AGA-BP neural network can effectively predict the concentration distribution of bioaerosols, and the predicted concentrations of bioaerosols are 1793 particles × m−3, 3088 particles × m−3, 5261 particles × m−3, 7410 particles × m−3 and 9133 particles × m−3 for air quality with superior, fine, mild contamination, middle level pollution and heavy pollution at an altitude of 0.315 km, respectively. We found that the predicted concentration of pollution weather is much higher than that of good weather. Furthermore, the AGA-BP neural network was used to predict the concentration profiles of atmospheric bioaerosols under different weather conditions, which provided a new research method for forecasting and alert of atmospheric bioaerosols.

Memory-based gradient-guided progressive propagation network for video deblurring

Memory-based gradient-guided progressive propagation network for video deblurring

Remora Optimized Dynamic Neural Network to Design Low-Power CNTFET Logic Gate Realization with Multiple-Valued Circuits

The emerging trend of new digital technology leads to the requirement of small, faster, and low-power consumed processors. The advanced 5 G technology forces the usage of enhanced battery life, maximized spectral efficiency, etc., these can be achieved by the inclusion of a Carbon nanotube field-effect transistor (CNTFET). The proposed multiple-valued logic circuits (MVL) are designed using some of the unique features of the CNTFET such as high transconductance, setting of desired threshold voltage, and similar mobility of P and N-type devices. For designing low-power and optimized logic gate realization we have utilized the inherent binary gates along with the CMOS architecture. Further, an artificial intelligence approach known as Dynamic Neural Network (DNN) is adopted for the realization of logic gates in the MVL with the employment of lesser MVL operators. To enhance the parameter utilization of DNN with the help of tuning, we exploited the Remora Optimization algorithm (ROA). The results are obtained by conducting experiments on HFSS software. This approach is designed to reduce the static power dissipation, the number of logic gates, and network propagation delay. Compared to the existing methods, the proposed technique outlined a standard voltage of 32 nm that considers 0.9 V supply voltage. The result of our proposed method can be deviated by only 25% and thus outperforms all the other approaches. The standby power consumption of proposed approach is 2.12 W

Modeling and adaptive control of demand oscillation propagation in an uncertain aerial transportation network

Modeling and adaptive control of demand oscillation propagation in an uncertain aerial transportation network

A fractional stochastic SPEIQR epidemic model in switching network for COVID-19

The rapid spread of COVID-19 globally is the cause of panic and billions of dollars in losses and some public policies have been implemented to contain the spread of the virus. It is highly significant to establish mathematical models to describe the spread of the pandemic. In this paper, a fractional stochastic susceptible–protected–exposed–infected–quarantined–recovered (SPEIQR) epidemic model in switching network is proposed to investigate COVID-19 and other diseases in the future, in which the switching connections in the propagation network of viruses based on a Bernoulli random variable are given to represent the policy change during COVID-19. The effects of environmental fluctuations are also considered by introducing the multiplicative noise terms into the disease transmission coefficient to obtain a more realistic mathematical model. Furthermore, the Riemann–Liouville fractional derivative is used to describe the dynamics of disease transmission process which is dependent on the history information of the stochastic system. Some qualitative properties of the model are derived and some sufficient conditions are obtained to ensure the local extinction with probability one of the disease. Finally, we use the epidemic data of three countries (Italy, Germany and Switzerland) and three cities in China (Hubei, Hunan and Henan) from January 22 to July 20 to demonstrate the validity of the derived results and the availability of the proposed models. This work provides some theoretical results to deeply understand the dynamics of COVID-19 and regulate the disease dynamics.

Swarm intention identification via dynamic distribution probability image

In the realm of decision-making for defense and security applications, it is paramount to swiftly and accurately identify the intentions of incoming swarms. Conventional identification methods predominantly focus on single-target applications and overlook the perturbations introduced by measurement noise. In this study, we propose a novel concept: the Dynamic Distribution Probability (DDP) image, which is constructed using the estimated state and its covariance matrix. Each grayscale pixel value within the image signifies the probability of the presence of the agent within the swarm. Our proposed identification scheme integrates the use of Extended Kalman Filter (EKF), Convolutional Neural Network (CNN), Back Propagation (BP) network, and Gated Recurrent Unit (GRU) network. Specifically, the DDP image is processed through a CNN to distill the formation characteristics, and the estimated swarm state from EKF is inputted into a BP network to deduce the kinematic information. The outputs from both networks are summed and subsequently channeled into a GRU network to capture temporal dynamics. Extensive numerical simulations and flight experiments are presented to demonstrate the robust anti-noise capability of the proposed scheme compared with conventional methods, as well as its superior training efficiency.

Development of CASRock for modeling multi-fracture interactions in rocks under hydro-mechanical conditions

The interaction between multiple fractures is important in the analysis of rock fracture propagation, fracture network evolution and stability and integrity of rocks under hydro-mechanical (HM) coupling conditions. At present, modeling the mechanical behavior of multiple fractures is still challenging. Under the condition of multiple fractures, the opening, closing, sliding, propagation and penetration of fractures become more complicated. In order to simulate the HM coupling behavior of multi-fracture system, the paper presents a novel numerical scheme, including mesh reconstruction and topology generation algorithm, to efficiently and accurately represent fractures and their propagation process, and a potential function-based algorithm to address contact problem. The fracture contact algorithm does not need to set contact pairs and thus is suitable for complex contact situations from small to large deformations induced by HM loading. The topology of fracture interfaces is constructed by the dynamic adding algorithm, which makes the mesh reconstruction more rapid in the modeling of fracturing process, especially in the case of multiple fractures intersections. The numerical scheme is implemented in CASRock, a self-developed numerical code, to simulate the propagation process of rock fractures and the interaction of multiple fractures under the condition of HM coupling. To verify the suitability of the code, a series of tests were performed. The code was then applied to simulate hydraulic fracture propagation and fracture interactions caused by fluid injection. The ability of this method to study fracture propagation, multi-fracture interaction and fracture network evolution under hydro-mechanical coupling conditions is demonstrated.

FAST-CA: Fusion-based Adaptive Spatial–Temporal Learning with Coupled Attention for airport network delay propagation prediction

The issue of delay propagation prediction in airport networks has garnered increasing global attention, particularly due to its profound impact on operational efficiency and passenger satisfaction in modern air transportation systems. Despite research advancements in this domain, existing methodologies often fall short of comprehensively addressing the challenges associated with predicting delay propagation in airport networks, especially in terms of handling complex spatial–temporal dependencies and sequence couplings. In response to the complex challenge of predicting delay propagation in airport networks, we introduce the Fusion-based Adaptive Spatial–Temporal Learning with Coupled Attention (FAST-CA) framework. FAST-CA is an innovative model that integrates dynamic and adaptive graph learning, coupled attention mechanisms, periodicity feature extraction, and multifaceted information fusion modules. This holistic approach enables a thorough analysis of the interplay between flight departure and arrival delays and the spatial–temporal correlations within airport networks. Rigorously evaluated on two extensive real-world datasets, our model consistently outperforms current state-of-the-art baseline models, showcasing superior predictive performance and the effective learning capabilities of its intricately designed modules. Our research highlights the criticality of analyzing spatial–temporal relationships and the dynamics of flight coupling, offering significant theoretical and practical contributions to the advancement and management of air transportation systems.

The impact of Caputo-Fabrizio fractional derivative and the dynamics of noise on worm propagation in wireless IoT networks

The main objective of this article on Wireless sensor network of the Internet of Things (IoT). The wireless network, B bluetooth network, infrared network, and other networks are the key components of the Internet of Things (IoT). The major emphasis of this work was on the impact of Caputo-Fabrizio fractional derivative on worm propagation in heterogeneous susceptible-exposed-infected-recovered wireless IoT devices. We first determined the equilibrium points and fundamental reproduction number for the Caputo-Fabrizio HSEIR system, and then we discussed the stability of the system at the worm propagation equilibrium point. Using the Picard-Lindeof method, we determine the existence and unique solution for the fractional CF system of the heterogeneous SEIR model. Next, we use fixed point theory to judge the stability of the iterative method. We investigate the impact of the derivative order on the behaviour of the resultant functions and acquired numerical values by computing the model's findings for various fractional orders. In addition, we compute the integer-order model's results and contrast them with the results of the fractional-order model. We develop a periodically intermittent controller driven by white noise with the amazing benefits of reduced cost and more adaptable control technique to restrict the spread of worms in wireless IoT networks. To clearly define the conditions for stability in probability one, we employ the stochastic analysis technique. Our results show that the nonlinear worm propagation system may be stabilised by intermittent stochastic perturbation under the parameters of intermittent time linked to stochastic perturbation strength. Our theoretical conclusions may be used to analyse the observable processes of the worm, design countermeasures to prevent its spread, and evaluate the consequences of various system parameters.

Analysis of Compressive Strength of Sustainable Fibre Reinforced Foamed Concrete Using Machine Learning Techniques

Abstract This study emphasizes the usage of Silica Fume (SF) and Marble Sludge Powder (MSP) as a partial replacement for fly ash in Fibre Reinforced Foamed Concrete (FRFC). The compressive strength for various samples was analyzed using Artificial Neural Network (ANN) methods. In this research work, the utilization of silica fume, fly ash, marble sludge powder, polypropylene fiber, and foaming agent in fiber-reinforced foamed concrete is presented and a sincere attempt has been made to use silica fume and marble sludge powder for the replacement of fly ash with various percentages. In addition to that polypropylene fiber (PPF) was used in various proportions of 0%, 0.1%, 0.2%, 0.3%, 0.4%, and 0.5%. The Feed Forward Propagation (FFP) network of the machine learning method with one hidden layer was taken as the ANN structure of FRFC. In this ANN work, cement, silica fume, fly ash, marble sludge powder, foaming agent, water, and polypropylene fiber were used as input parameters and compressive strength is the output parameter. The correlation coefficient with the ANN methods was found as 0.940 for compressive strength. In machine learning techniques, the ANN method was found to be accurate in estimating and analyzing strength prediction responses with effective parameters.

On propagation in networks, promising models beyond network diffusion to describe degenerative brain diseases and traumatic brain injuries.

Introduction: Connections among neurons form one of the most amazing and effective network in nature. At higher level, also the functional structures of the brain is organized as a network. It is therefore natural to use modern techniques of network analysis to describe the structures of networks in the brain. Many studies have been conducted in this area, showing that the structure of the neuronal network is complex, with a small-world topology, modularity and the presence of hubs. Other studies have been conducted to investigate the dynamical processes occurring in brain networks, analyzing local and large-scale network dynamics. Recently, network diffusion dynamics have been proposed as a model for the progression of brain degenerative diseases and for traumatic brain injuries. Methods: In this paper, the dynamics of network diffusion is re-examined and reaction-diffusion models on networks is introduced in order to better describe the degenerative dynamics in the brain. Results: Numerical simulations of the dynamics of injuries in the brain connectome are presented. Different choices of reaction term and initial condition provide very different phenomenologies, showing how network propagation models are highly flexible. Discussion: The uniqueness of this research lies in the fact that it is the first time that reaction-diffusion dynamics have been applied to the connectome to model the evolution of neurodegenerative diseases or traumatic brain injury. In addition, the generality of these models allows the introduction of non-constant diffusion and different reaction terms with non-constant parameters, allowing a more precise definition of the pathology to be studied.

Preventing malware propagation in wireless sensor networks: Hybrid optimization algorithm for controlling

Malware transmission is a significant security issue in WSN, however, the influence of the attack and defensive processes on malware propagation is rarely taken into account in traditional malware propagation prevention methods. Advanced methods are in need to stop the propagation of malware of sensor nodes. With the formulation of representing dynamics among states, a new decision-making problem as the optimal control problem via hybrid optimization algorithm. The proposing model is termed as Butterfly Updated Bald Eagle Optimization based Prevention of Malware Propagation in Wireless Sensor Network (BUBEO-PMPWSN). In the proposed controlling system, optimal system parameters are analyzed via the BUBEO for preventing malware propagation in WSN. Particularly, the sensor node states considered are Susceptible, Infectious, Infectious and sleeping, recovered, Recovered and sleeping, and finally Dead. The system parameter tuning will be under the evaluation of fitness calculation under probability of infectious sensor node becoming recovered and the probability of infectious sensor node entering sleeping state. This optimal tuning strategy ensures the preventing of malware propagation. Finally, the performance of proposed BUBEO-PMPWSN model is evaluated and validated successfully by comparing other state-of-the-art models. The BUBEO-PMPWSN achieved 250 recovered nodes for time 500, while the HGS, BOA, HBA, COOT, and HHO scored 123, 115, 236, 172, and 180, respectively, for recovered nodes.

Prediction of hydrological and water quality data based on granular-ball rough set and k-nearest neighbor analysis.

Hydrological and water quality datasets usually encompass a large number of characteristic variables, but not all of these significantly influence analytical outcomes. Therefore, by wisely selecting feature variables with rich information content and removing redundant features, it not only can the analysis efficiency be improved, but the model complexity can also be simplified. This paper considers introducing the granular-ball rough set algorithm for feature variable selection and combining it with the k-nearest neighbor method and back propagation network to analyze hydrological and water quality data, thus promoting overall and fused inspection. The results of hydrological water quality data analysis show that the proposed method produces better results compared to using a standalone k-nearest neighbor regressor.

Lithium response in bipolar disorder is associated with focal adhesion and PI3K-Akt networks: a multi-omics replication study

Lithium is the gold standard treatment for bipolar disorder (BD). However, its mechanism of action is incompletely understood, and prediction of treatment outcomes is limited. In our previous multi-omics study of the Pharmacogenomics of Bipolar Disorder (PGBD) sample combining transcriptomic and genomic data, we found that focal adhesion, the extracellular matrix (ECM), and PI3K-Akt signaling networks were associated with response to lithium. In this study, we replicated the results of our previous study using network propagation methods in a genome-wide association study of an independent sample of 2039 patients from the International Consortium on Lithium Genetics (ConLiGen) study. We identified functional enrichment in focal adhesion and PI3K-Akt pathways, but we did not find an association with the ECM pathway. Our results suggest that deficits in the neuronal growth cone and PI3K-Akt signaling, but not in ECM proteins, may influence response to lithium in BD.

Relation fusion propagation network for transductive few-shot learning

Previous graph-based meta-learning approaches have explored pairwise feature similarity to learn instance-level relations of samples, however, the gap between the sample relations in feature and label spaces is often ignored. It is empirically observed that instances with different labels may display considerable similarity in visual characteristics, making it challenging to distinguish between them in the feature space. To this end, we propose a dual-branch Relation Fusion Propagation Network (RFPN) for transductive few-shot learning, which explicitly models both feature and label relations across support-query pairs. Specifically, we design a Relation Fusion Block (RFB) to fuse instance-level and class-level relations, thus obtaining more robust fusion-level relations to guide feature propagation. In addition to the feature propagation branch, we encode the label relations and construct a label shortcut branch for label propagation. To alleviate the error accumulation during propagation, which is caused by uncertain pseudo-labels for query samples, we propose a Label Shortcut Mechanism (LSM) to progressively update the sample relations with the initial labels. Our full method is plug-and-play and can be easily applied in existing graph-based approaches for transductive few-shot learning. Extensive experiments demonstrate that our proposed RFPN yields significant improvements over the baselines and achieves promising performance on four popular few-shot classification benchmarks.

Interplay of network topologies in aviation delay propagation: A complex network and machine learning analysis

In this study, the fundamental characteristics of flight delay propagation and the key factors influencing such propagation are investigated. Three distinct types of networks were constructed: an aviation network, a traffic flow network, and a delay propagation network. Employing complex network theory, an analysis of the fundamental topological attributes of each network was conducted, exploring the interrelations among these attributes. Findings reveal a notable resemblance between the network topology attributes of the delay propagation network and the traffic flow network. The delay propagation network overall adheres to scale-free network attributes, with a few dominant major airports playing a pivotal role in the delay propagation process. Furthermore, it was uncovered that smaller airports are more susceptible to the influence of delay propagation than their larger counterparts. Delays, in a general sense, exhibit a pronounced tendency to aggregate and spread extensively within the realm of smaller airports. Regarding delay propagation, airport traffic emerges as the primary factor precipitating this phenomenon, with a feature importance score reaching 0.889. An escalation in airport traffic significantly amplifies the extent of delay propagation, yet concurrently, this escalation is not incessant but stabilizes beyond a certain threshold.

Joint learning of structural and textual information on propagation network by graph attention networks for rumor detection

Joint learning of structural and textual information on propagation network by graph attention networks for rumor detection