General Network Research Articles

Learning to Search a Lightweight Generalized Network for Medical Image Fusion

Learning to Search a Lightweight Generalized Network for Medical Image Fusion

DM-Fusion: Deep Model-Driven Network for Heterogeneous Image Fusion.

Heterogeneous image fusion (HIF) is an enhancement technique for highlighting the discriminative information and textural detail from heterogeneous source images. Although various deep neural network-based HIF methods have been proposed, the most widely used single data-driven manner of the convolutional neural network always fails to give a guaranteed theoretical architecture and optimal convergence for the HIF problem. In this article, a deep model-driven neural network is designed for this HIF problem, which adaptively integrates the merits of model-based techniques for interpretability and deep learning-based methods for generalizability. Unlike the general network architecture as a black box, the proposed objective function is tailored to several domain knowledge network modules to model the compact and explainable deep model-driven HIF network termed DM-fusion. The proposed deep model-driven neural network shows the feasibility and effectiveness of three parts, the specific HIF model, an iterative parameter learning scheme, and data-driven network architecture. Furthermore, the task-driven loss function strategy is proposed to achieve feature enhancement and preservation. Numerous experiments on four fusion tasks and downstream applications illustrate the advancement of DM-fusion compared with the state-of-the-art (SOTA) methods both in fusion quality and efficiency. The source code will be available soon.

Повышение энергоэффективности предприятия за счет установки в системах его водоснабжения гидроэлектрических энергосистем по выработке электроэнергии

The article presents the results of the research strategy for the development of mountainous territories of Kazakhstan. The study of the ossibilities of developing the energy system of mountainous territories is shown on the example of the city of Almaty (Kazakhstan). Materials and Methods. To study the energy efficiency indicators of enterprises in the manufacturing industry, Almaty Heavy Machinery Plant (AHMP) was selected as one of the largest machine-building enterprises in Almaty. Results. It is shown that in the conditions of supplying the water supply system with industrial water from mountain rivers, hydro turbines can be used. It was found that such hydro turbines can be placed in water supply pipelines of production shops. Discussion. An assessment was made that showed that the electricity generated by them can be used to power their own equipment, as well as to supply electricity to the general network. Conclusion. The obtained average efficiency of the hydro turbine of 39 % allows us to calculate the possible generation of electricity in real conditions up to 11 kW, while the declared capacity was 16 kW. Resume. If we consider the tasks of energy saving in production, we can confidently predict that the generation of electricity from the developed hydro turbine will reduce the total electricity consumed from the grid. Suggestions for practical applications and directions for future research. The mini-hydroelectric power station developed and presented in this work can be used in water supply systems of industrial enterprises located in mountainous regions. This system can contribute to increasing the energy efficiency of enterprises. In the future, it is planned to scale these mini-hydroelectric power stations and make smaller and larger units for use in larger and smaller enterprises.

Maximin headway control of automated vehicles for system optimal dynamic traffic assignment in general networks

This study develops the headway control framework in a fully automated road network, as we believe headway of Automated Vehicles (AVs) is another influencing factor to traffic dynamics in addition to conventional vehicle behaviors (e.g. route and departure time choices). Specifically, we aim to search for the optimal time headway between AVs on each link that achieves the network-wide system optimal dynamic traffic assignment (SO-DTA). To this end, the headway-dependent fundamental diagram (HFD) and headway-dependent double queue model (HDQ) are developed to model the effect of dynamic headway on roads, and a dynamic network model is built. It is rigorously proved that the minimum headway could always achieve SO-DTA, yet the optimal headway is non-unique. Motivated by these two findings, this study defines a novel concept of maximin headway, which is the largest headway that still achieves SO-DTA in the network. Mathematical properties regarding maximin headway are analyzed and an efficient solution algorithm is developed. Numerical experiments on both a small and large network verify the effectiveness of the maximin headway control framework as well as the properties of maximin headway. This study sheds light on deriving the desired solution among the non-unique solutions in SO-DTA and provides implications regarding the safety margin of AVs under SO-DTA.

Computation of ultra-short-term prediction intervals of the power prosumption in active distribution networks

Microgrids and, in general, active distribution networks require ultra-short-term prediction, i.e., for sub-second time scales, for specific control decisions. Conventional forecasting methodologies are not effective at such time scales. To address this issue, we propose a non-parametric method for computing ultra short-term prediction intervals (PIs) of the power prosumption of generic electrical-distribution networks. The method groups historical observations into clusters according to the values of influential variables. It is applied either to the original or to the differentiated power-prosumption time series. The clusters are considered statistically representative pools of future realizations of power prosumption (or its derivative). They are used to determine empirical PDFs and, by extracting the quantiles, to deliver PIs for respective arbitrary confidence levels. The models are validated a posteriori by carrying out a performance analysis that uses experimentally observed power-prosumption for different building types, thus allowing the identification of the dominant model.

Modern trends in the formation of sports and recreational complexes

Today, in the context of the demographic and environmental crisis in Ukraine caused by military actions, it is extremely important to consider the trend of formation of sports infrastructure facilities as those subjects of the urban recreational network that will take the main position in the post-war period in the task of rehabilitation of the population. The nature of the urban structure makes us look at the modern sports complex as an integrated entity of the general recreational network. As a result, the problem of interconnection of the sports function with other functions in the structure of a single multifunctional complex is becoming more relevant. The article highlights the problems of formation of sports and recreational complexes in modern conditions. The factors that influence the modification of the form of SRC and the vectors of further development of the typology of this type of buildings and structures are identified. The design of sports infrastructure facilities should include a generalisation of modern experience and the application of the method of prospective analysis to determine trends in the development of the form of the SRC. This involves analysing current needs and forecasting future requirements for sports and recreational facilities, taking into account changes in public attitudes, demographic trends and the impact of technological innovations. The study is based on a systematic approach where sports and recreational complexes are considered as an element of the socio-cultural space. The use of a holistic architectural approach involved the application of methods for generalising modern experience in the design of facilities of this type and the method of prospective analysis, which will help to identify trends in the development of the form of sports and recreation complexes. The results of the study can be used to develop and adjust the existing architectural research base and the strategy for the reconstruction of Ukraine. They will contribute to the creation of an effective, modern and sustainable sports infrastructure in modern Ukrainian cities.

Generic network sparsification via degree- and subgraph-based edge sampling

Network (or graph) sparsification accelerates many downstream analyses. For graph sparsification, sampling methods derived from local heuristic considerations are common in practice, due to their efficiency in generating sparse subgraphs using only local information. Filtering-based edge sampling is the most typical approach in this respect, yet it heavily depends on an appropriate definition of edge importance. Instead, we propose a generalized node-focused edge sampling framework by preserving scaled/expected local node characteristics. Apart from expected degrees, these local node characteristics include the expected number of triangles and the expected number of non-closed wedges associated with a node. From a technical point of view, we adapt a game-theoretic sampling method from uncertain graph generation to obtain sparse subgraphs that approximate the expected local properties. We include a tolerance threshold for much faster convergence. Within this framework, we provide appropriate algorithmic variants for sparsification. Moreover, we propose a network measure called tri-wedge assortativity for the selection of the most suitable variant when sparsifying a given network. Extensive experimental studies on functional climate, observed real-world, and synthetic networks show the effectiveness of our method in preserving overall structural network properties – on average consistently better than the state of the art.

Cancer Patients' Social Relationships During 3 Years After Diagnosis-Generic and Cancer-Specific Social Networks.

Social relationships are important health resources and may be investigated as social networks. We measured cancer patients' social subnetworks divided into generic social networks (people known to the patients) and disease-specific social networks (the persons talked to about the cancer) during 3years after diagnosis. Newly diagnosed patients with localized breast cancer (n = 222), lymphoma (n = 102), and prostate cancer (n = 141) completed a questionnaire on their social subnetworks at 2-5months after diagnosis and 9, 18, and 36months thereafter. Generic and cancer-specific numbers of persons of spouse/partner; other family; close relatives, in detail; and friends were recorded as well as cancer-specific numbers of persons in acquaintances; others with cancer; work community; healthcare professionals; and religious, hobby, and civic participation. The data was analyzed with regression models. At study entry, most patients had a spouse/partner, all had close relatives (the younger, more often parents; and the older, more often adult children with families) and most also friends. The cancer was typically discussed with them, and often with acquaintances and other patients (74-86%). Only minor usually decreasing time trends were seen. However, the numbers of distant relatives and friends were found to strongly increase by the 9-month evaluation (P < 0.001). Cancer patients have multiple social relationships and usually talk to them about their cancer soon after diagnosis. Most temporal changes are due to the natural course of life cycle. The cancer widened the patients' social networks by including other patients and healthcare professionals and by an increased number of relatives and friends.

Magnitude and angle dynamics in training single ReLU neurons

Understanding the training dynamics of deep ReLU networks is a significant area of interest in deep learning. However, there remains a lack of complete elucidation regarding the weight vector dynamics, even for single ReLU neurons. To bridge this gap, our study delves into the training dynamics of the gradient flow w(t) for single ReLU neurons under the square loss, dissecting it into its magnitude ‖w(t)‖ and angle φ(t) components. Through this decomposition, we establish upper and lower bounds on these components to elucidate the convergence dynamics. Furthermore, we demonstrate the empirical extension of our findings to general two-layer multi-neuron networks. All theoretical results are generalized to the gradient descent method and rigorously verified through experiments.

Auditory Sensory Gating: Effects of Noise.

Cortical auditory evoked potentials (CAEPs) indicate that noise degrades auditory neural encoding, causing decreased peak amplitude and increased peak latency. Different types of noise affect CAEP responses, with greater informational masking causing additional degradation. In noisy conditions, attention can improve target signals' neural encoding, reflected by an increased CAEP amplitude, which may be facilitated through various inhibitory mechanisms at both pre-attentive and attentive levels. While previous research has mainly focused on inhibition effects during attentive auditory processing in noise, the impact of noise on the neural response during the pre-attentive phase remains unclear. Therefore, this preliminary study aimed to assess the auditory gating response, reflective of the sensory inhibitory stage, to repeated vowel pairs presented in background noise. CAEPs were recorded via high-density EEG in fifteen normal-hearing adults in quiet and noise conditions with low and high informational masking. The difference between the average CAEP peak amplitude evoked by each vowel in the pair was compared across conditions. Scalp maps were generated to observe general cortical inhibitory networks in each condition. Significant gating occurred in quiet, while noise conditions resulted in a significantly decreased gating response. The gating function was significantly degraded in noise with less informational masking content, coinciding with a reduced activation of inhibitory gating networks. These findings illustrate the adverse effect of noise on pre-attentive inhibition related to speech perception.

A mathematical framework for measuring and tuning tempo in developmental gene regulatory networks.

Embryo development is a dynamic process governed by the regulation of timing and sequences of gene expression, which control the proper growth of the organism. Although many genetic programmes coordinating these sequences are common across species, the timescales of gene expression can vary significantly among different organisms. Currently, substantial experimental efforts are focused on identifying molecular mechanisms that control these temporal aspects. In contrast, the capacity of established mathematical models to incorporate tempo control while maintaining the same dynamical landscape remains less understood. Here, we address this gap by developing a mathematical framework that links the functionality of developmental programmes to the corresponding gene expression orbits (or landscapes). This unlocks the ability to find tempo differences as perturbations in the dynamical system that preserve its orbits. We demonstrate that this framework allows for the prediction of molecular mechanisms governing tempo, through both numerical and analytical methods. Our exploration includes two case studies: a generic network featuring coupled production and degradation, with a particular application to neural progenitor differentiation; and the repressilator. In the latter, we illustrate how altering the dimerisation rates of transcription factors can decouple the tempo from the shape of the resulting orbits. We conclude by highlighting how the identification of orthogonal molecular mechanisms for tempo control can inform the design of circuits with specific orbits and tempos.

UVS-CNNs: Constructing general convolutional neural networks on quasi-uniform spherical images

Omnidirectional images, also known as spherical images, offer a significant advantage for the environmental sensing of mobile robots due to their wide field of view. However, previous studies of constructing convolutional neural networks on spherical images have been limited by non-uniform pixel sampling, leading to suboptimal performance in semantic segmentation. To address this issue, a novel pixel segmentation approach is proposed to achieve near-uniform pixel distribution across the entire spherical surface. The corresponding convolution operation for the resulting image is designed as well, which extends the capabilities of spherical CNNs from semantic segmentation to more complex tasks such as instance segmentation. The method is evaluated on the Stanford 2D3DS dataset and shows superior performance compared to conventional spherical CNNs. Furthermore, the method also achieves impressive instance segmentation results on our experimental LiDAR data, demonstrating the general feasibility of our approach for common CNN tasks. The related code and dataset are released in the following link: https://github.com/YoungRainy/UVS-U-Net.

A generalized method for in-plane eigenproblem of a cable network attached with multiple rigid cross-ties

Considering the cable sag as well as the impacts of cables’ dynamic axial forces on the transverse forces, this paper makes an effort to propose a generalized method for studying in-plane free vibration of the cable network attached with multiple rigid cross-ties. First, the vibration equations of the suspended cables are established and solved by separation-of-variables method. Then, the system is divided into several substructures, which are reassembled according to transfer matrix method (TMM). Using the boundary conditions, the characteristic equation of the system is yielded. In this way, the basic theory for solving the eigenproblem of a general cable network attached with rigid cross-ties is derived. Four examples are given to illustrate the applicability of the current method, and their frequencies and mode shapes are acquired by a developed calculating program. Meanwhile, a finite element model (FEM) is also established. It is found that the results are in a good agreement with those obtained by FEM, which verifies the validity and correctness of the method in this paper.

SAILoR: Structure-Aware Inference of Logic Rules.

Boolean networks provide an effective mechanism for describing interactions and dynamics of gene regulatory networks (GRNs). Deriving accurate Boolean descriptions of GRNs is a challenging task. The number of experiments is usually much smaller than the number of genes. In addition, binarization leads to a loss of information and inconsistencies arise in binarized time-series data. The inference of Boolean networks from binarized time-series data alone often leads to complex and overfitted models. To obtain relevant Boolean models of gene regulatory networks, inference methods could incorporate data from multiple sources and prior knowledge in terms of general network structure and/or exact interactions. We propose the Boolean network inference method SAILoR (Structure-Aware Inference of Logic Rules). SAILoR incorporates time-series gene expression data in combination with provided reference networks to infer accurate Boolean models. SAILoR automatically extracts topological properties from reference networks. These can describe a more general structure of the GRN or can be more precise and describe specific interactions. SAILoR infers a Boolean network by learning from both continuous and binarized time-series data. It navigates between two main objectives, topological similarity to reference networks and correspondence with gene expression data. By incorporating the NSGA-II multi-objective genetic algorithm, SAILoR relies on the wisdom of crowds. Our results indicate that SAILoR can infer accurate and biologically relevant Boolean descriptions of GRNs from both a static and a dynamic perspective. We show that SAILoR improves the static accuracy of the inferred network compared to the network inference method dynGENIE3. Furthermore, we compared the performance of SAILoR with other Boolean network inference approaches including Best-Fit, REVEAL, MIBNI, GABNI, ATEN, and LogBTF. We have shown that by incorporating prior knowledge about the overall network structure, SAILoR can improve the structural correctness of the inferred Boolean networks while maintaining dynamic accuracy. To demonstrate the applicability of SAILoR, we inferred context-specific Boolean subnetworks of female Drosophila melanogaster before and after mating.

A generic causality-informed neural network (CINN) methodology for quantitative risk analytics and decision support.

In this paper, we develop a generic framework for systemically encoding causal knowledge manifested in the form of hierarchical causality structure and qualitative (or quantitative) causal relationships into neural networks to facilitate sound risk analytics and decision support via causally-aware intervention reasoning. The proposed methodology for establishing causality-informed neural network (CINN) follows a four-step procedure. In the first step, we explicate how causal knowledge in the form of directed acyclic graph (DAG) can be discovered from observation data or elicited from domain experts. Next, we categorize nodes in the constructed DAG representing causal relationships among observed variables into several groups (e.g., root nodes, intermediate nodes, and leaf nodes), and align the architecture of CINN with causal relationships specified in the DAG while preserving the orientation of each existing causal relationship. In addition to a dedicated architecture design, CINN also gets embodied in the design of loss function, where both intermediate and leaf nodes are treated as target outputs to be predicted by CINN. In the third step, we propose to incorporate domain knowledge on stable causal relationships into CINN, and the injected constraints on causal relationships act as guardrails to prevent unexpected behaviors of CINN. Finally, the trained CINN is exploited to perform intervention reasoning with emphasis on estimating the effect that policies and actions can have on the system behavior, thus facilitating risk-informed decision making through comprehensive "what-if" analysis. Two case studies are used to demonstrate the substantial benefits enabled by CINN in risk analytics and decisionsupport.

Multiple feedback recurrent neural network based super-twisting predefined-time nonsingular terminal sliding mode control for quad-rotor UAV

This research deals with the overall predefined-time stability (PTS) of Unmanned Aerial Vehicles (UAV) ensuring rapid convergence based on a novel sliding mode control (SMC). The strength of predefined-time sliding manifolds lies in the convergence rate can be adjusted by an explicit parameter. For the limitation of chattering encountered by predefined-time SMC (PTSMC), a variable gain super-twisting algorithm (STA) with additional linear items is designed as the switch controller. To conserve the restrained computational resources of quadrotors, the equivalent control input is approximated by a multiple feedback recurrent neural network (MFRNN) directly, which is challenging for general recurrent neural networks. The proposed MFRNN is characterized by the incorporation of double-loop feedback within the layers, augmenting its capacity for accurate approximation. To address the vanishing gradients commonly encountered with traditional activation functions, LeakyRelu is chosen. The Lyapunov theory is utilized to ensure the PTS of overall system and obtain the MFRNN weight update laws. An experiment is conducted to validate the proposed scheme.

FDCNN-AS: Federated deep convolutional neural network Alzheimer detection schemes for different age groups

Alzheimer's disease (AD) is a memory-related disease that occurs in the human brain where neurons become degenerative. It is an evolved form of dementia that deteriorates over time. Machine learning, an extended version of deep learning, has appeared as an optimistic strategy for AD detection. Regardless, the existing AD detection approaches have yet to acquire the expected accuracy, mainly due to unreasonable data for training and testing. In this paper, we present the Federated Deep Convolutional Neural Network Alzheimer Detection Schemes (FDCNN-AS), specifically designed for varying age groups. FDCNN-AS is an efficient framework that contains architecture, algorithm flow, and implementation. It manages AD data from various laboratories and processes it in additional clinics. Our method mixes training data models from different types of data such as positron emission tomography, summed tomography, magnetic resonance imaging, blood tests, and questionnaires about synaptic degeneration. Further, we look at some restrictions that have yet to be addressed in AD detection. These include seeing AD at different ages, extrapolating the severity of brain damage, comparing treatment and recovery rates, and finding benign and malignant ranges in AD data that has been collected. To ensure secure and privacy-preserving learning, we execute FDCNN-AS within a federated learning environment that concerns considerable laboratories and clinics. Within this setup, we operate the generic deep convolutional neural network. The experimental results indicate that FDCNN-AS performs optimally, reaching a remarkable 99% accuracy in detecting dementia Alzheimer's in the human brain.

Effects of network heterogeneity on phases of the quenched contact process in directed complex networks

We investigate the quenched contact process within directed complex networks, including directed random networks and directed scale-free networks. Additionally, we explore the contact process within directed general small-world networks and directed hierarchical modular networks. We observed the presence of a Griffiths phase characterized by continuously varying critical exponents of the order parameter. Notably, the directionality of the networks plays a significant role in influencing the Griffiths phase within the quenched contact process. We suggest that the presence of strongly directed connecting clusters is responsible for the changes in the continuously varying critical exponents of the order parameter within directed networks.

A novel formulation of low voltage distribution network equivalents for reliability analysis

Reliability analysis of large power networks requires accurate aggregate models of low voltage (LV) networks to allow for reasonable calculation complexity and to prevent long computational times. However, commonly used lumped load models neglect the differences in spatial distribution of demand, type of phase-connection of served customers and implemented protection system components (e.g., single-pole vs three-pole). This paper proposes a novel use of state enumeration (SE) and Monte Carlo simulation (MCS) techniques to formulate more accurate LV network reliability equivalents. The combined SE and MCS method is illustrated using a generic suburban LV test network, which is realistically represented by a reduced number of system states. This approach allows for a much faster and more accurate reliability assessments, where further reduction of system states results in a single-component equivalent reliability model with the same unavailability as the original LV network. Both mean values and probability distributions of standard reliability indices are calculated, where errors associated with the use of single-line models, as opposed to more detailed three-phase models, are quantified.

Is human perception reliable? Toward illumination robust food freshness prediction from food appearance — Taking lettuce freshness evaluation as an example

Machine vision-based food quality evaluation systems have achieved great attention in industry and academia in recent years because of their high-throughput and non-invasive properties. In practice, the environmental illumination conditions variations would cause visual evaluation bias for both human and machine perceptions. Compared to other existing studies which take sample images and human visual grading under fixed illumination conditions, this study first investigated the environmental illumination effects on the performance of visual-based food grading for both humans and machines. Taking lettuce samples as an example, an image dataset that considered the environmental illumination variations was first established. This dataset encompasses human visual grading scores obtained from sensory panels as well. In contrast to current studies that utilize a single grader or L*a*b* color features to assess sample freshness, our study reveals a substantial impact of environmental illumination on both human perception (p<0.0001) and L*a*b* color features (p<0.0001). In order to enhance the performance of machine vision-based freshness prediction, multitask learning protocols were incorporated into the proposed new network architectures. This allowed the simultaneous prediction of both sample freshness and illumination conditions. In comparison to commonly used generic convolutional neural network models and vision transformer models, the newly proposed model exhibited superior freshness prediction performance. It minimized the prediction error by 20.36%, outperforming the generic ResNet model. This research represents the first quantitative study addressing human and machine perceptions under varied illumination conditions for food quality evaluation. The findings are anticipated to play a pivotal role in expediting the integration of machine vision applications into food engineering practices.