Search

Physics-informed neural networks (PINNs) have been widely used to solve partial differential equations in recent years. But studies have shown that there is a gradient pathology in PINNs. That is, there is an imbalance gradient problem in each regularization term during back-propagation, which makes it difficult for neural network models to accurately approximate partial differential equations. Based on the depth-weighted residual neural network and neural attention mechanism, we propose a new mixed-weighted residual block in which the weighted coefficients are chosen autonomously by the optimization algorithm, and one of the transformer networks is replaced by a skip connection. Finally, we test our algorithms with some partial differential equations, such as the non-homogeneous Klein–Gordon equation, the (1+1) advection–diffusion equation, and the Helmholtz equation. Experimental results show that the proposed algorithm significantly improves the numerical accuracy. [ABSTRACT FROM AUTHOR]

Non-ferrous metals are important bulk commodities and play a significant part in the development of society. Their price forecast is of great reference value for investors and policymakers. However, developing a robust price forecast model is tricky due to the price's drastic fluctuations. In this work, a novel fusion model based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Singular Spectrum Analysis (SSA), and Long Short-Term Memory (LSTM) is constructed for non-ferrous metals price forecast. Considering the complexity of their price change, the dual-stage signal preprocessing which combines CEEMDAN and SSA is utilized. Firstly, we use the CEEMDAN algorithm to decompose the original nonlinear price sequence into multiple Intrinsic Mode Functions (IMFs) and a residual. Secondly, the component with maximum sample entropy is decomposed by SSA; this is the so-called Multivariate Mode Decomposition (MMD). A series of experimental results show that the proposed MMD-LSTM method is more stable and robust than the other seven benchmark models, providing a more reasonable scheme for the price forecast of non-ferrous metals. [ABSTRACT FROM AUTHOR]

In this paper, block neural network (BNN) method is proposed to solve several kinds of differential equations. BNN is used to construct approximating functions and its derivatives, the improved extreme learning machine (IELM) algorithm are designed to train network weights. To evaluate the performance of the proposed method, numerical examples are performed by the presented method. Comparison of the numerical results with exact solutions validate the feasibility of the proposed method in accuracy. Results compared with other recent research works also validate the superiority of the proposed approach. Numerical results show that the proposed BNN with IELM algorithm perform well in accuracy and requires less hidden neurons. [ABSTRACT FROM AUTHOR]

To provide a new numerical algorithm for solving elliptic partial differential equations (PDEs), the Legendre neural network (LNN) and improved extreme learning machine (IELM) algorithm are introduced to propose a Legendre improved extreme learning machine (L-IELM) method, which is applied to solving elliptic PDEs in this paper. The product of two Legendre polynomials is chosen as basis functions of hidden neurons. Single hidden layer LNN is used to construct approximate solutions and its derivatives of differential equations. IELM algorithm is used for network weights training, and the algorithm steps of the proposed L-IELM method are summarized. Finally, in order to evaluate the present algorithm, various test examples are selected and solved by the proposed approach to validate the calculation accuracy. Comparative study with the earlier methods in literature is described to verify the superiority of the presented L-IELM method. Experiment results show that the proposed L-IELM algorithm can perform well in terms of accuracy and execution time, which in addition provides a new algorithm for solving elliptic PDEs. [ABSTRACT FROM AUTHOR]

With the increasing demands for vast amounts of data and high-speed signal transmission, the use of multi-conductor transmission lines is becoming more common. The impact of transmission lines on signal transmission is thus a key issue affecting the performance of high-speed digital systems. To solve the problem of lossless two-conductor transmission line equations (LTTLEs), a neural network model and algorithm are explored in this paper. By selecting the product of two triangular basis functions as the activation function of hidden layer neurons, we can guarantee the separation of time, space, and phase orthogonality. By adding the initial condition to the neural network, an improved extreme learning machine (IELM) algorithm for solving the network weight is obtained. This is different to the traditional method for converting the initial condition into the iterative constraint condition. Calculation software for solving the LTTLEs based on the IELM algorithm is developed. Numerical experiments show that the results are consistent with those of the traditional method. The proposed neural network algorithm can find the terminal voltage of the transmission line and also the voltage of any observation point. It is possible to calculate the value at any given point by using the neural network model to solve the transmission line equation. [ABSTRACT FROM AUTHOR]

In this paper, we study numerical solution of ruin probability of continuous time model. We develop an effective optimal adaptive particle swarm optimization-triangular neural network (PSO-TNN), which consists of three parts: particle swarm optimization algorithm (PSO) improved trigonometric function, extreme learning machine algorithm with initial conditions (IELM) and improved reduction algorithm. The results obtained that PSO-TNN is superior to triangular neural network (TNN) and physics-informed neural networks (PINNs), and PSO is superior to Aquila Optimizer (AO), Smell Agent Optimization (SAO), African vultures optimization algorithm (AVOA), Arithmetic optimization algorithm (AOA) in the optimization of neural network. Because the relationship between the number of neural networks and the mean square error is uncertain, we propose the adaptive reduction algorithm (AR). Through the comparison of numerical solutions with the analytical solutions and traditional numerical solutions, the PSO-TNN algorithm clearly reduced the mean square error and relative error. The PSO-TNN algorithm shows a clear improvement in terms of accuracy and overall efficiency. [ABSTRACT FROM AUTHOR]

The neural circuits controlling feeding and emotional behaviors are intricately and reciprocally connected. Recent technological developments, including cell type-specific optogenetic and chemogenetic approaches, allow functional characterization of genetically defined cell populations and neural circuits in feeding and emotional processes. Here we review recent studies that have utilized circuit-based manipulations to decipher the functional interactions between neural circuits controlling feeding and those controlling emotional processes. Specifically, we highlight newly described neural circuit interactions between classical emotion-related brain regions, such as the hippocampus and amygdala, and homeostatic feeding circuitry in the arcuate nucleus and lateral hypothalamus (LH). Together these circuits will provide a template for future studies to examine functional interactions between feeding and emotion. [ABSTRACT FROM AUTHOR]

Summary: Synaptic plasticity in response to changes in physiologic state is coordinated by hormonal signals across multiple neuronal cell types. Here, we combine cell-type-specific electrophysiological, pharmacological, and optogenetic techniques to dissect neural circuits and molecular pathways controlling synaptic plasticity onto AGRP neurons, a population that regulates feeding. We find that food deprivation elevates excitatory synaptic input, which is mediated by a presynaptic positive feedback loop involving AMP-activated protein kinase. Potentiation of glutamate release was triggered by the orexigenic hormone ghrelin and exhibited hysteresis, persisting for hours after ghrelin removal. Persistent activity was reversed by the anorexigenic hormone leptin, and optogenetic photostimulation demonstrated involvement of opioid release from POMC neurons. Based on these experiments, we propose a memory storage device for physiological state constructed from bistable synapses that are flipped between two sustained activity states by transient exposure to hormones signaling energy levels. [Copyright &y& Elsevier]

This paper develops a Legendre neural network method (LNN) for solving linear and nonlinear ordinary differential equations (ODEs), system of ordinary differential equations (SODEs), as well as classic Emden-Fowler equations. The Legendre polynomial is chosen as a basis function of hidden neurons. A single hidden layer Legendre neural network is used to eliminate the hidden layer by expanding the input pattern using Legendre polynomials. The improved extreme learning machine (IELM) algorithm is used for network weights training when solving algebraic equation systems, and several algorithm steps are summed up. Convergence was analyzed theoretically to support the proposed method. In order to demonstrate the performance of the method, various testing problems are solved by the proposed approach. A comparative study with other approaches such as conventional methods and latest research work reported in the literature are described in detail to validate the superiority of the method. Experimental results show that the proposed Legendre network with IELM algorithm requires fewer neurons to outperform the numerical algorithm in the latest literature in terms of accuracy and execution time. [ABSTRACT FROM AUTHOR]

Needle coke has attracted much attention as soft carbon anode material for potassium-ion batteries (PIBs) owing to its high crystallinity and conductivity. However, limited research has been conducted on the storage mechanism and the functional and structural evolution of the solid electrolyte interphase (SEI) in ester and ether based electrolytes. This work investigates the potassium ion storage mechanisms of soft carbon anodes in three electrolytes: KPF 6 EC/DEC, KPF 6 DME, and KFSI DME, and reveals that K+ stored in the KPF 6 EC/DEC electrolyte could be through an "adsorption-intercalation" mechanism. Whereas K+ stored in the DME-based electrolytes is believed to undergo a "co-intercalation/adsorption-intercalation" process. Interfacial analyses indicate that the SEI formed in the EC/DEC electrolyte inhibits co-intercalation of potassium ion and solvent, which preserves sufficient space to accommodate potassium ion. Conversely, the [K-DME]+ solvation co-intercalation in DME-based electrolytes compromises the structural integrity, and results in a persistent capacity decline. This work offers novel insights into potassium ion storage in ester and ether-based electrolytes, and provides theoretical support for the fabrication of high-performance soft carbon anodes. The storage mechanism and interfacial evolution of needle coke anodes for potassium-ion batteries were systematically investigated by using KPF 6 EC/DEC, KPF 6 DME, and KFSI DME electrolytes. [Display omitted] [ABSTRACT FROM AUTHOR]

2 rhMG53 lessens H2O2-induced oxidative injury to hUC-MSCs and promotes cell migration. a Representative images of hUC-MSCs with and without 200 M H2O2 treatment. b Time- and dose-dependent effects of H2O2 on hUC-MSCs. Two hundred micromolar H2O2 was used for subsequent experiments to induce hUC-MSC oxidative damage. c Dose-dependent effects of MG53 on hUC-MSCs. [Extracted from the article]

This study aims to investigate the relationship between DNA methylation and expression of human dopamine transporter ( hDAT ). We examined methylation status of hDAT in cells with various hDAT expression levels, including two dopaminergic neural cell lines (SK-N-AS and SH-SY-5Y) and one non-dopaminergic cell line (HEK293) by bisulfite sequencing PCR(BSP). The effects of DNA methyltransferase inhibitor 5-aza-dC or/and histone deacetylase inhibitor (HDACi, sodium butyrate, NaB) on the DNA methylation status and mRNA expression levels of hDAT were examined. The results revealed marked hypomethylation of the two promoter regions (−1214 to −856 bp and −48 to 439 bp, the first base of exon 1 was taken as +1 bp)of hDAT in SK-N-AS (4.7% ± 2.0 mC and 3.5% ± 1.0 mC , respectively) compared with SH-SY-5Y (88.0% ± 4.4% mC and 81.1% ± 8.8% mC ) and HEK293 (90.7% ± 2.4 mC and 84.4% ± 8.6% mC ) cell lines, indicating a cell-specific methylation regulation of hDAT . 5-aza-dC and NaB decreased hypermethylation,while increase hDAT expression in SH-SY-5Y cells and recovered hDAT mRNA expression in HEK293 cells. DNA methylation enabled the cell-specific differential expression of the hDAT gene. hDAT silencing was reversed by the introduction of DNA hypomethylation via 5-aza-dC or/and NaB. [ABSTRACT FROM AUTHOR]

Objective: It is well established that schizophrenia patients have high cardiovascular morbidity and mortality. However, the underlying risk factors in the earliest stages of both schizophrenia illness and antipsychotics treatment are less clear. This study aimed to characterize the metabolic features of those patients.Methods: We performed a retrospective cohort study in a naturalistic setting, which included antipsychotic-naïve, first-episode schizophrenia (FES) inpatients with the baseline metabolic measurements and changes following a short term treatment with antipsychotic drugs.Results: Although prevalence of hypertriglyceridemia, hypercholesterolemia, higher-LDL-C and hyperglycaemia in patients with FES were much lower than those of the general population (7.5% v.s. 14.9%, 9.2% v.s. 18.4%, 8.1% v.s. 14.9%, 8.6% v.s.18.3%, respectively), lower-HDL-C in patients with FES were much more prevalent than that of the general population (19.9% v.s. 6.4%). Despite significant metabolic risk profiles (as such lipid abnormalities and insulin resistance) increase, mean fasting glucose and glucosylated serum protein (GSP) were significantly decreased after the short term (median of 23days) antipsychotics exposure, compared to baseline. There is no significant difference of the metabolic profile change between monopharmacy and polypharmacy.Conclusion: These results indicated an early-onset nature of HDL-C abnormalities in drug-naïve FES patients. Lipids metabolism risk may develop early and quickly after antipsychotic exposure. Early monitoring is required for the purpose of early detection and hence prevention of the initial metabolic risk which may lead to diabetes mellitus and cardiovascular disease. [ABSTRACT FROM AUTHOR]

The objective of this study was to determine the effects of different doses of caffeine on appetite and anxiety-related behavior. Additionally, we sought to determine if withdrawal from chronic caffeine administration promotes anxiety. In this study, we utilized rodent open field testing and feeding behavior assays to determine the effects of caffeine on feeding and anxiety-related behavior (n = 8 mice; 4-8 weeks old). We also measured 2 h and 24 h food intake and body-weight during daily administration of caffeine (n = 12 mice; 4-8 weeks old). To test for caffeine withdrawal induced anxiety, anxiety-related behavior in rodents was quantified following withdrawal from four consecutive days of caffeine administration (n = 12 mice; 4-8 weeks old). We find that acute caffeine administration increases food intake in a dose-dependent manner with lower doses of caffeine more significantly increasing food intake than higher doses. Acute caffeine administration also reduced anxiety-related behaviors in mice without significantly altering locomotor activity. However, we did not observe any differences in 24 h food intake or body weight following chronic caffeine administration and there were no observable differences in anxiety-related behaviors during caffeine withdrawal. In conclusion, we find that caffeine can both increase appetite and decrease anxiety-related behaviors in a dose dependent fashion. Given the complex relationship between appetite and anxiety, the present study provides additional insights into potential caffeine-based pharmacological mechanisms governing appetite and anxiety disorders, such as bulimia nervosa. [ABSTRACT FROM AUTHOR]

Ample evidence indicates that obesity causes dysfunctions in the lung. Previous studies also show that cholinergic anti-inflammatory pathways play crucial roles in obesity-induced chronic inflammation via α7 nicotinic acetylcholine receptor (α7nAChR) signaling. However, it remains unclear whether and how obesity affects the expressions of α7nAChR in myeloid cells in the lung. To address this question, we treated regular chow diet-fed mice or high-fat diet induced obese mice with lipopolysaccharide (LPS) or vehicle via endotracheal injections. By using a multicolor flow cytometry approach to analyze and characterize differential cell subpopulations and α7nAChR expressions, we find no detectable α7nAChR in granulocytes, monocytes and alveolar macrophages, and low expression levels of α7nAChR were detected in interstitial macrophages. Interestingly, we find that a challenge with LPS treatment significantly increased expression levels of α7nAChR in monocytes, alveolar and interstitial macrophages. Meanwhile, we observed that the expression levels of α7nAChR in alveolar and interstitial macrophages in high-fat diet induced obese mice were lower than regular chow diet-fed mice challenged by the LPS. Together, our findings indicate that obesity alters the expressions of α7nAChR in differential lung myeloid cells. [ABSTRACT FROM AUTHOR]

Background: Traumatic brain injury (TBI) is a common neurotrauma leading to brain dysfunction and death. Human umbilical cord-derived mesenchymal stem cells (hUC-MSCs) hold promise in the treatment of TBI. However, their efficacy is modest due to low survival and differentiation under the harsh microenvironment of the injured brain. MG53, a member of TRIM family protein, plays a vital role in cell and tissue damage repair. The present study aims to test whether MG53 preserves hUC-MSCs against oxidative stress and enhances stem cell survival and efficacy in TBI treatment. Methods: In this study, we performed a series of in vitro and in vivo experiments in hUC-MSCs and mice to define the function of MG53 enhancing survival, neurogenesis, and therapeutic efficacy of stem cells in murine traumatic brain injury. Results: We found that recombinant human MG53 (rhMG53) protein protected hUC-MSCs against H2O2-induced oxidative damage and stimulated hUC-MSC proliferation and migration. In a mouse model of contusion-induced TBI, intravenous administration of MG53 protein preserved the survival of transplanted hUC-MSCs, mitigated brain edema, reduced neurological deficits, and relieved anxiety and depressive-like behaviors. Co-treatment of MG53 and hUC-MSCs enhanced neurogenesis by reducing apoptosis and improving PI3K/Akt-GSK3β signaling. Conclusion: MG53 enhances the efficacy of hUC-MSCs in the recovery of TBI, indicating that such adjunctive therapy may provide a novel strategy to lessen damage and optimize recovery for brain injury. [ABSTRACT FROM AUTHOR]

This paper introduces the improved LS-SVM algorithms for solving two-point and multi-point boundary value problems of high-order linear and nonlinear ordinary differential equations. To demonstrate the reliability and powerfulness of the improved LS-SVM algorithms, some numerical experiments for third-order, fourth-order linear and nonlinear ordinary differential equations with two-point and multi-point boundary conditions are performed. The idea can be extended to other complicated ordinary differential equations. [ABSTRACT FROM AUTHOR]