Simple Network Model Research Articles

Using generic cities to assess flying ad-hoc network (FANET) performance

Abstract Flying ad-hoc networks (FANET) offer a solution to provide communication among unmanned aerial systems (UAS) without ground infrastructure and, hence, may provide redundancy if main terrestrial communication systems fail. With a strongly growing UAS market, this redundancy is of particular importance especially if the vehicles are operated over densely populated regions, such as urban areas. To further develop the FANET technology, it is, therefore, essential to understand and predict the performance requirements for such networks and the necessary systems as they arise from the different applications. However, with the diversity of cities, the problem arises that custom-made simulation models often lack generality. For this reason, in our work, we introduce an approach to generate randomized cities and to model UAS traffic for selected applications to assess different performance metrics of the FANET. This approach comprises the creation of randomized city geometries, the derivation of UAS traffic demand on these geometries, as well as the simulation of the flights of each individual vehicle. To showcase the capabilities of the approach, we generate an exemplary city and obtain characteristic datasets of the UAS traffic which are then used to assess the FANET performance for a selected FANET application using a simplified network model. We will show that FANET performance is strongly influenced by the extent of the demand, demand distribution over time, vehicle speed, and delivery network geometry.

Detection of Domain Name Server Amplification Distributed Reflection Denial of Service Attacks Using Convolutional Neural Network-Based Image Deep Learning

Domain Name Server (DNS) amplification Distributed Reflection Denial of Service (DRDoS) attacks are a Distributed Denial of Service (DDoS) attack technique in which multiple IT systems forge the original IP of the target system, send a request to the DNS server, and then send a large number of response packets to the target system. In this attack, it is difficult to identify the attacker because of its ability to deceive the source, and unlike TCP-based DDoS attacks, it usually uses the UDP protocol, which has a fast communication speed and amplifies network traffic by simple manipulating options, making it one of the most widely used DDoS techniques. In this study, we propose a simple convolutional neural network (CNN) model that is designed to detect DNS amplification DRDoS attack traffic and has hyperparameters adjusted through experiments. As a result of evaluating the accuracy of the proposed CNN model for detecting DNS amplification DRDoS attacks, the average accuracy of the experiment was 0.9995, which was significantly better than several machine learning (ML) models in terms of performance. It also showed good performance compared to other deep learning (DL) models, and, in particular, it was confirmed that this simple CNN had the fastest time in terms of execution compared to other deep learning models by experimentation.

Simple and Efficient Equivariant Message-Passing Neural Network Model for Non-local Potential Energy Surfaces.

Machine learning potential has become increasingly successful in atomistic simulations. Many of these potentials are based on an atomistic representation in a local environment, but an efficient description of nonlocal interactions that exceed a common local environment remains a challenge. Herein, we propose a simple and efficient equivariant model, EquiREANN, to effectively represent a nonlocal potential energy surface. It relies on a physically inspired message-passing framework, where the fundamental descriptors are linear combinations of atomic orbitals, while both invariant orbital coefficients and the equivariant orbital functions are iteratively updated. We demonstrate that this EquiREANN model is able to describe the subtle potential energy variation due to the nonlocal structural change with high accuracy and little extra computational cost than an invariant message passing model. Our work offers a generalized approach to create equivariant message-passing adaptations of other advanced local many-body descriptors.

On inequalities of shear modulus contributions in disordered elastic bodies

Abstract We investigate generic inequalities of various contributions to the shear modulus $\mu$ in ensembles of amorphous elastic bodies. We focus first on a simple elastic network model with connectivity matrices (CMs) which are either annealed or quenched, at or out of equilibrium. The stress-fluctuation formalism relation for $\mu$ is rewritten as $\mu = \mu_1 + \mu_a$ with $\mu_1 \ge 0$ characterizing the variance of the quenched shear stresses and $\mu_a$ being a simple average over all states and CMs. For equilibrium CM-distributions $\mu_a$ becomes equivalent to the shear modulus of annealed systems, i.e. $\mu_a \ge 0$, while more generally $\mu_a$ may become strongly negative as shown by considering a temperature quench and a scalar active two-temperature model. Consistent relations are also found for glass-forming colloids where $\mu-\mu_1=\mu_a=0$ for equilibrium ensembles, i.e. $\mu$ is set by the quenched shear stresses, while $\mu_a$ becomes again negative otherwise.

Neural network modelling of kinematic and dynamic features for signature verification

Online signature parameters, which are based on human characteristics, broaden the applicability of an automatic signature verifier. Although kinematic and dynamic features have previously been suggested, accurately measuring features such as arm and forearm torques remains challenging. We present two approaches for estimating angular velocities, angular positions, and force torques. The first approach involves using a physical UR5e robotic arm to reproduce a signature while capturing those parameters over time. The second method, a cost-effective approach, uses a neural network to estimate the same parameters. Our findings demonstrate that a simple neural network model can extract effective parameters for signature verification. Training the neural network with the MCYT300 dataset and cross-validating with other databases, namely, BiosecurID, Visual, Blind, OnOffSigDevanagari-75 and OnOffSigBengali-75 confirm the model’s generalization capability. The trained model is available at: https://github.com/gvessio/SignatureKinematics.

The spectrum of experimentally accessible states for melting and freezing transitions in mesoporous materials.

Structural disorder in mesoporous solids gives rise to complex phase behavior for materials confined within their pore spaces. As a result, a wide spectrum of possible phase configurations associated with spatial distributions of thermodynamic phases throughout the pore networks can be realized in experiments. Despite their importance, quantifying these states remains largely unaddressed. By considering solid-liquid equilibria as a representative example and using a simple random network model, we investigate the spectrum of such states accessible in real experiments and relate this spectrum to the structural characteristics of porous solids. We classify these states by their free energies and demonstrate how network effects break degeneracies for specific phase compositions and temperatures. Furthermore, we identify the experimental conditions that delineate boundary free energy states, differentiating accessible from inaccessible states. The insights from this study on solid-liquid equilibria are also equally applicable to gas-liquid equilibria in confined spaces and contribute to a deeper understanding of relaxation dynamics associated with hysteresis.

Analysis of the Hyperparameter Selection in Machine Learning

Abstract. A Machine learning is now playing a role in various fields. With the development of the internet, the volume of data is increasing, and the algorithms of machine learning are becoming more complex. Adjusting hyperparameters is a challenge for beginners who are new to machine learning. This research implemented a simple neural network model through code. Moreover, this study has changed some of the most basic parameters of the model and trained different models. This study has used visualization methods to show how the model training time and performance change with the alteration of hyperparameters, and derived their respective accuracy rates. This article provides entry-level data and code for newcomers to the field of machine learning. Since this experiment uses a federated learning model, employs local_update SGD, and also utilizes parallel methods, it is quite comprehensive, covering many algorithm codes for machine learning beginners. Testing the basic parameters of the model can help newcomers to machine learning understand more clearly the impact of hyperparameters on the model. It also aids in the development of more algorithms for adjusting hyperparameters.

The role of motor cortex in motor sequence execution depends on demands for flexibility.

The role of the motor cortex in executing motor sequences is widely debated, with studies supporting disparate views. Here we probe the degree to which the motor cortex's engagement depends on task demands, specifically whether its role differs for highly practiced, or 'automatic', sequences versus flexible sequences informed by external cues. To test this, we trained rats to generate three-element motor sequences either by overtraining them on a single sequence or by having them follow instructive visual cues. Lesioning motor cortex showed that it is necessary for flexible cue-driven motor sequences but dispensable for single automatic behaviors trained in isolation. However, when an automatic motor sequence was practiced alongside the flexible task, it became motor cortex dependent, suggesting that an automatic motor sequence fails to consolidate subcortically when the same sequence is produced also in a flexible context. A simple neural network model recapitulated these results and offered a circuit-level explanation. Our results critically delineate the role of the motor cortex in motor sequence execution, describing the conditions under which it is engaged and the functions it fulfills, thus reconciling seemingly conflicting views about motor cortex's role in motor sequence generation.

Temporal resolution of spike coding in feedforward networks with signal convergence and divergence.

Convergent and divergent structures in the networks that make up biological brains are found across many species and brain regions at various spatial scales. Neurons in these networks fire action potentials, or "spikes", whose precise timing is becoming increasingly appreciated as large sources of information about both sensory input and motor output. While previous theories on coding in convergent and divergent networks have largely neglected the role of precise spike timing, our model and analyses place this aspect at the forefront. For a suite of stimuli with different timescales, we demonstrate that structural bottlenecks- small groups of neurons post-synaptic to network convergence - have a stronger preference for spike timing codes than expansion layers created by structural divergence. Additionally, we found that a simple network model based on convergence and divergence ratios of a hawkmoth (Manduca sexta) nervous system can reproduce the relative contribution of spike timing information in its motor output, providing testable predictions on optimal temporal resolutions of spike coding across the moth sensory-motor pathway at both the single-neuron and population levels. Our simulations and analyses suggest a relationship between the level of convergent/divergent structure present in a feedforward network and the loss of stimulus information encoded by its population spike trains as their temporal resolution decreases, which could be tested experimentally across diverse neural systems in future studies. We further show that this relationship can be generalized across different spike-generating models and measures of coding capacity, implying a potentially fundamental link between network structure and coding strategy using spikes.

A Parametric Study on NOx Emissions From Ammonia Containing Product Gas in Rich–Quench–Lean Combustion

Abstract Due to climate change, there has been an increasing demand for fuels that can accelerate the transition away from fossil fuels to clean energy. Humidified product gas obtained from gasifying biomass is emerging as a promising candidate to replace natural gas, as it is composed of a gaseous mixture of hydrogen, steam, carbon monoxide, and methane. However, the gasification process releases ammonia and other nitrogen bearing compounds into the product gas, resulting in substantial increases in nitric oxides, NOx, in the exhaust. As such, there has been a recent push to understand the underlying chemical kinetics that drive NOx formation in order to optimize gas turbines to mitigate emissions at the source. In this study, a simplified chemical reactor network (CRN) model for a gas turbine rich–quench–lean (RQL) combustor was developed in cantera. The following parameters were investigated in this study: equivalence ratio of the primary section, overall equivalence ratio, steam dilution, postflame residence time, and recirculation from the postquench region to the primary section. Additionally, a benchmark CRN representing a lean burner (LB) is also developed. Results of the CRN model suggest that, when comparing to LB, a RQL type combustor delivers up to a 75% reduction in emissions. Additionally, it was found that, for both the LB and RQL combustors, an overall lean to stoichiometric equivalence ratio is well suited to reduce emissions in highly humidified fuels, while for moderately humidified fuels it is preferable to operate in an overall slightly rich equivalence ratio. The difference observed is mainly due to the fact that, at high humidification and lean conditions, the temperature is favorable for the conversion of ammonia to nitrogen, while, at moderate humidification and rich conditions, NO reacts with ammonia in the reburn process. Finally, it is suggested that the incorporation of recirculation from the secondary section to the primary section of the RQL burner results in a broader low emission region, due to more favorable conditions for ammonia conversion to nitrogen in the primary section.

Roles of network topology in the relaxation dynamics of simple chemical reaction network models

Understanding the relationship between the structure of chemical reaction networks and their reaction dynamics is essential for unveiling the design principles of living organisms. However, while some network-structural features are known to relate to the steady-state characteristics of chemical reaction networks, mathematical frameworks describing the links between out-of-steady-state dynamics and network structure are still underdeveloped. Here, we characterize the out-of-steady-state behavior of a class of artificial chemical reaction networks consisting of the ligation and splitting reactions of polymers. Within this class, we examine minimal networks that can convert a given set of sources (e.g., nutrients) to a specified set of targets (e.g., biomass precursors). By exploring the dynamics of the models with a simple setup, we find three distinct types of relaxation dynamics after perturbation from a steady-state: exponential-, power-law-, and plateau-dominated. We computationally show that we can predict this out-of-steady-state dynamical behavior from just three features computed from the network’s stoichiometric matrix, namely, (1) the rank gap, determining the existence of a steady-state; (2) the left null-space, being related to conserved quantities in the dynamics; and (3) the stoichiometric cone, dictating the range of achievable chemical concentrations. We further demonstrate that these three quantities relates to the type of relaxation dynamics of combinations of our minimal networks, larger networks with many redundant pathways, and a real example of a metabolic network. The relationship between the topological features of reaction networks and the relaxation dynamics presented here are useful clues for understanding the design of metabolic reaction networks as well as industrially useful chemical production pathways.

Evaluation Approach and Controller Design Guidelines for Subsequent Commutation Failure in Hybrid Multi-Infeed HVDC System

Due to the difference in output characteristics between the line-commutated converter-based high-voltage direct current (LCC-HVDC) and voltage-source converter-based high-voltage direct current (VSC-HVDC), the hybrid multi-infeed high-voltage direct current (HMIDC) presents complex coupling characteristics. As the AC side is disturbed, the commutation failure (CF) occurring on the LCC side is the main factor threatening the safe operation of the system. In this paper, the simplified equivalent network model of HMIDC is established by analyzing the output characteristics of VSC and LCC. Hereafter, based on the derived model and the control system of LCC-HVDC, the dynamic equations of the extinction angle are deduced. Consequently, by applying the phase portrait method, the causes of CF occurring in the HMIDC system as well as the impacts of control parameters on the transient stability are revealed. Furthermore, the stabilization boundaries for the reference value of the DC voltage are obtained via the above analysis. Finally, the theoretical analysis is verified by the simulations in the PSCAD/EMTDC.

Generating Daily High-Resolution Regional XCO2 by Deep Neural Network and Multi-Source Data

CO2 is one of the primary greenhouse gases impacting global climate change, making it crucial to understand the spatiotemporal variations of CO2. Currently, commonly used satellites serve as the primary means of CO2 observation, but they often suffer from striping issues and fail to achieve complete coverage. This paper proposes a method for constructing a comprehensive high-spatiotemporal-resolution XCO2 dataset based on multiple auxiliary data sources and satellite observations, utilizing multiple simple deep neural network (DNN) models. Global validation results against ground-based TCCON data demonstrate the excellent accuracy of the constructed XCO2 dataset (R is 0.94, RMSE is 0.98 ppm). Using this method, we analyze the spatiotemporal variations of CO2 in China and its surroundings (region: 0°–60° N, 70°–140° E) from 2019 to 2020. The gapless and fine-scale CO2 generation method enhances people’s understanding of CO2 spatiotemporal variations, supporting carbon-related research.

Of criminals and cancer: The importance of social bonds and innate morality on cellular societies

The current dogma in cancer biology contends that cancer is an identity problem: mutations in a cell's DNA cause it to “go rogue” and proliferate out of control. However, this largely ignores the role of cell-cell interaction and fails to explain phenomena such as cancer reversion, the existence of cancers without mutations, and foreign-body carcinogenesis. In this proof-of-concept paper, we draw on criminology to propose that cancer may alternatively be conceptualized as a relational problem: Although a cell's genetics is essential, the influence of its interaction with other cells is equally important in determining its phenotype. We create a simple agent-based network model of interactions among normal and cancer cells to demonstrate this idea. We find that both high mutation rates and low levels of connectivity among cells can promote oncogenesis. Viewing cancer as a breakdown in communication networks among cells in a tissue complements the gene-centric paradigm nicely and provides a novel perspective for understanding and treating cancer.

A Neural Network Model for Efficient Musculoskeletal-Driven Skin Deformation

We present a comprehensive neural network to model the deformation of human soft tissues including muscle, tendon, fat and skin. Our approach provides kinematic and active correctives to linear blend skinning [Magnenat-Thalmann et al. 1989] that enhance the realism of soft tissue deformation at modest computational cost. Our network accounts for deformations induced by changes in the underlying skeletal joint state as well as the active contractile state of relevant muscles. Training is done to approximate quasistatic equilibria produced from physics-based simulation of hyperelastic soft tissues in close contact. We use a layered approach to equilibrium data generation where deformation of muscle is computed first, followed by an inner skin/fascia layer, and lastly a fat layer between the fascia and outer skin. We show that a simple network model which decouples the dependence on skeletal kinematics and muscle activation state can produce compelling behaviors with modest training data burden. Active contraction of muscles is estimated using inverse dynamics where muscle moment arms are accurately predicted using the neural network to model kinematic musculotendon geometry. Results demonstrate the ability to accurately replicate compelling musculoskeletal and skin deformation behaviors over a representative range of motions, including the effects of added weights in body building motions.

Deep learning with convolution neural network detecting mesiodens on panoramic radiographs: comparing four models.

The aim of this study was to develop an optimal, simple, and lightweight deep learning convolutional neural network (CNN) model to detect the presence of mesiodens on panoramic radiographs. A total of 628 panoramic radiographs with and without mesiodens were used as training, validation, and test data. The training, validation, and test dataset were consisted of 218, 51, and 40 images with mesiodens and 203, 55, and 61 without mesiodens, respectively. Unclear panoramic radiographs for which the diagnosis could not be accurately determined and other modalities were required for the final diagnosis were retrospectively identified and employed as the training dataset. Four CNN models provided within software supporting the creation of neural network models for deep learning were modified and developed. The diagnostic performance of the CNNs was evaluated according to accuracy, precision, recall and F1 scores, receiver operating characteristics (ROC) curves, and area under the ROC curve (AUC). In addition, we used SHapley Additive exPlanations (SHAP) to attempt to visualize the image features that were important in the classifications of the model that exhibited the best diagnostic performance. A binary_connect_mnist_LeNet model exhibited the best performance of the four deep learning models. Our results suggest that a simple lightweight model is able to detect mesiodens. It is worth referring to AI-based diagnosis before an additional radiological examination when diagnosis of mesiodens cannot be made on unclear images. However, further revaluation by the specialist would be also necessary for careful consideration because children are more radiosensitive than adults.

On the evolutionary origin of discrete phenotypic plasticity.

Phenotypic plasticity provides an attractive strategy for adapting to various environments, but the evolutionary mechanism of the underlying genetic system is poorly understood. We use a simple gene regulatory network model to explore how a species acquires phenotypic plasticity, particularly focusing on discrete phenotypic plasticity, which has been difficult to explain by quantitative genetic models. Our approach employs a population genetic framework that integrates the developmental process, where each individual undergoes growth to develop its phenotype, which subsequently becomes subject to selection pressures. Our model considers two alternative types of environments, with the gene regulatory network including a sensor gene that turns on and off depending on the type of environment. With this assumption, we demonstrate that the system gradually adapts by acquiring the ability to produce two distinct optimum phenotypes under two types of environments without changing genotype, resulting in phenotypic plasticity. We find that the resulting plasticity is often discrete after a lengthy period of evolution. Our results suggest that gene regulatory networks have a notable capacity to flexibly produce various phenotypes in response to environmental changes. This study also shows that the evolutionary dynamics of phenotype may differ significantly between mechanistic-based developmental models and quantitative genetics models, suggesting the utility of incorporating gene regulatory networks into evolutionary models.

Designing Decentralized Systems with High Survivability Inspired by Altruistic Social Interactions of Vampire Bats

Altruism is a key concept in the design of decentralized systems with high survivability. We focus on a community of vampire bats to reveal how intra-group altruism produces group-wide survivability. Although these bats die within three days if food is unavailable, they can survive for over 10 years by developing a highly sophisticated social community in which they share food. This food-sharing behavior occurs not only among blood relatives, but also among unrelated individuals through self-organizing social relationships based on grooming behavior. We propose a simple network model that focuses on the relationship between food sharing and grooming. We performed simulations under periodic, stationary, and irregular feeding environments, and found that suitable update rules for social relationships depend on the type of environment. Our findings provide insights into how decentralized systems with high survivability can be designed based on altruism.

An end-to-end method for predicting compound-protein interactions based on simplified homogeneous graph convolutional network and pre-trained language model

Identification of interactions between chemical compounds and proteins is crucial for various applications, including drug discovery, target identification, network pharmacology, and elucidation of protein functions. Deep neural network-based approaches are becoming increasingly popular in efficiently identifying compound-protein interactions with high-throughput capabilities, narrowing down the scope of candidates for traditional labor-intensive, time-consuming and expensive experimental techniques. In this study, we proposed an end-to-end approach termed SPVec-SGCN-CPI, which utilized simplified graph convolutional network (SGCN) model with low-dimensional and continuous features generated from our previously developed model SPVec and graph topology information to predict compound-protein interactions. The SGCN technique, dividing the local neighborhood aggregation and nonlinearity layer-wise propagation steps, effectively aggregates K-order neighbor information while avoiding neighbor explosion and expediting training. The performance of the SPVec-SGCN-CPI method was assessed across three datasets and compared against four machine learning- and deep learning-based methods, as well as six state-of-the-art methods. Experimental results revealed that SPVec-SGCN-CPI outperformed all these competing methods, particularly excelling in unbalanced data scenarios. By propagating node features and topological information to the feature space, SPVec-SGCN-CPI effectively incorporates interactions between compounds and proteins, enabling the fusion of heterogeneity. Furthermore, our method scored all unlabeled data in ChEMBL, confirming the top five ranked compound-protein interactions through molecular docking and existing evidence. These findings suggest that our model can reliably uncover compound-protein interactions within unlabeled compound-protein pairs, carrying substantial implications for drug re-profiling and discovery. In summary, SPVec-SGCN demonstrates its efficacy in accurately predicting compound-protein interactions, showcasing potential to enhance target identification and streamline drug discovery processes.Scientific contributionsThe methodology presented in this work not only enables the comparatively accurate prediction of compound-protein interactions but also, for the first time, take sample imbalance which is very common in real world and computation efficiency into consideration simultaneously, accelerating the target identification and drug discovery process.

Designing the geometry of compact tension specimens for easy fracture toughness measurement using reinforcement learning

Abstract For composite laminates, a rising R-curve is observed for their fracture toughness under Mode I stress, which is important for a comprehensive failure analysis of the materials. Since it is laborious to measure the R-curve due to its dependence on both the load and the crack extension, we put forward a novel compact tension specimen by modifying its geometry to eliminate the relation between fracture toughness and crack extension, so as to simplify the experimental process of the R-curve measurement by only recording the load history. Two machine learning models were developed for the optimum sample design based on the finite element analysis of the effect of sample geometries on the R-curve. A simple neural network model was built for designing tapered specimen and a reinforcement learning model was created for further finding the best design from a broader design space. The results showed that, in contrast to the specimens with a tapered shape, which only ensure the independence between the R-curve and crack extension in the case of a small extension, the design provided by the reinforcement learning provides such independence across a wider range of crack length and an improved accuracy.