Multi-scale problems in Computational Fluid Dynamics(CFD)often require numerous simulations across various design parameters.Using a fixed mesh for all cases may fail to capture critical physical features.Moving mesh ...Multi-scale problems in Computational Fluid Dynamics(CFD)often require numerous simulations across various design parameters.Using a fixed mesh for all cases may fail to capture critical physical features.Moving mesh adaptation provides an optimal resource allocation to obtain high-resolution flow-fields on low-resolution meshes.However,most existing methods require manual experience and the flow posteriori information poses great challenges to practical applications.In addition,generating adaptive meshes directly from design parameters is difficult due to highly nonlinear relationships.The diffusion model is currently the most popular model in generative tasks that integrates the diffusion principle into deep learning to capture the complex nonlinear correlations.A dual diffusion framework,Para2Mesh,is proposed to predict the adaptive meshes from design parameters by exploiting the robust data distribution learning ability of the diffusion model.Through iterative denoising,the proposed dual networks accurately reconstruct the flow-field to provide flow features as supervised information,and then achieve rapid and reliable mesh movement.Experiments in CFD scenarios demonstrate that Para2Mesh predicts similar meshes directly from design parameters with much higher efficiency than traditional method.It could become a real-time adaptation tool to assist engineering design and optimization,providing a promising solution for high-resolution flow-field analysis.展开更多
The formamide-titanium oxide interaction mechanism is a research target of great importance for understanding the elementary events of the origin of life: the synthesis of nucleoside bases and formation of biological ...The formamide-titanium oxide interaction mechanism is a research target of great importance for understanding the elementary events of the origin of life: the synthesis of nucleoside bases and formation of biological molecules needed for life. Titanium oxide (TiO2) can act as a strongly adsorbing surface or a catalytic material. In the present study, a comparative molecular dynamics analysis performed to clarify the adsorbing and diffusion properties of liquid formamide on a TiO2 surface in the presence of water molecules. The structural features of the formamide concentration effect (the accumulation of molecules) on a TiO2 surface in the presence and absence of water solvent are cleared up. Modification of the formamide diffusion abilities mediated by a water solvent is observed to correlate with the formamide-water concentration distribution on the surface.展开更多
Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizin...Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizing graph diffusion and contrastive learning.DGCL_DWA first employs personalized PageRank to generate a diffusion graph,revealing hidden biological connections.Chebyshev graph convolution extracts features from both the PPI and diffusion networks,and neighborhood contrastive learning harmonizes gene representations,reducing noise.The network-specific features are refined via Chebyshev graph convolutions,which are constrained via node classification and link prediction.A dynamic weight adjustment strategy balances task-specific losses during training.Finally,logistic regression is used to predict driver genes.The experimental results demonstrate the superior performance in pan-cancer and specific cancer driver gene identification compared with state-of-the-art methods.Ablation studies confirm the positive impact of the diffusion graph,contrastive learning,and dynamic weight adjustment on predictive accuracy.The source codes are available at https://doi.org/10.57760/sciencedb.31933.展开更多
Locating the source of diffusion in complex networks is a critical and challenging problem,exemplified by tasks such as identifying the origin of power grid faults or detecting the source of computer viruses.The accur...Locating the source of diffusion in complex networks is a critical and challenging problem,exemplified by tasks such as identifying the origin of power grid faults or detecting the source of computer viruses.The accuracy of source localization in most existing methods is highly dependent on the number of infected nodes.When there are few infected nodes in the network,the accuracy is relatively limited.This poses a major challenge in identifying the source in the early stages of diffusion.This article presents a novel deep learning-based model for source localization under limited information conditions,denoted as GCN-MSL (Graph Convolutional Networks and network Monitor-based Source Localization model).The GCN-MSL model is less affected by the number of infected nodes and enables the efficient identification of the diffusion source in the early stages.First,pre-deployed monitor nodes,controlled by the network administrator,continuously report real-time data,including node states and the arrival time of anomalous signals.These data,along with the network topology,are used to construct node features.Graph convolutional networks are employed to aggregate information from multiple-order neighbors,thereby forming comprehensive node representations.Subsequently,the model is trained with the true source labeled as the target,allowing it to distinguish the source node from other nodes within the network.Once trained,the model can be applied to locate hidden sources in other diffusion networks.Experimental results across multiple data sets demonstrate the superiority of the GCN-MSL model,especially in the early stages of diffusion,where it significantly enhances both the accuracy and efficiency of source localization.Additionally,the GCN-MSL model exhibits strong robustness and adaptability to variations in external parameters of monitor nodes.The proposed method holds significant value in the timely detection of anomalous signals within complex networks and preventing the spread of harmful information.展开更多
Renewable energy production and the balance between production and demand have become increasingly crucial in modern power systems,necessitating accurate forecasting.Traditional deterministic methods fail to capture t...Renewable energy production and the balance between production and demand have become increasingly crucial in modern power systems,necessitating accurate forecasting.Traditional deterministic methods fail to capture the inherent uncertainties associated with intermittent renewable sources and fluctuating demand patterns.This paper proposes a novel denoising diffusion method for multivariate time series probabilistic forecasting that explicitly models the interdependencies between variables through graph modeling.Our framework employs a parallel feature extraction module that simultaneously captures temporal dynamics and spatial correlations,enabling improved forecasting accuracy.Through extensive evaluation on two world real-datasets focused on renewable energy and electricity demand,we demonstrate that our approach achieves state-of-the-art performance in probabilistic energy time series forecasting tasks.By explicitly modeling variable interdependencies and incorporating temporal information,our method provides reliable probabilistic forecasts,crucial for effective decision-making and resource allocation in the energy sector.Extensive experiments validate that our proposed method reduces the Continuous Ranked Probability Score(CRPS)by 2.1%-70.9%,Mean Absolute Error(MAE)by 4.4%-52.2%,and Root Mean Squared Error(RMSE)by 7.9%-53.4%over existing methods on two real-world datasets.展开更多
针对基于知识图谱的推荐系统中存在的高阶建模困难与用户特征建模不足的问题,提出基于多跳机制的扩散图谱推荐模型(a diffusion map recommendation model based on multi-hop mechanism,MultiHop-GDN)。该模型通过端到端方法挖掘知识...针对基于知识图谱的推荐系统中存在的高阶建模困难与用户特征建模不足的问题,提出基于多跳机制的扩散图谱推荐模型(a diffusion map recommendation model based on multi-hop mechanism,MultiHop-GDN)。该模型通过端到端方法挖掘知识图谱高阶语义信息,涵盖知识图谱构建、特征提取网络构建与多跳扩散模型构建三部分内容。利用用户特征和项目特征构建知识图谱;深入分析用户兴趣、偏好和历史行为等信息,构建用户画像和兴趣模型;提出特征提取网络捕获深层次语义信息,通过本文模型的计算得到预测值。在两个公开数据集的对比实验表明,MultiHop-GDN能够同时实现用户和项目的高阶建模,与其他代表论文的模型相比有良好的推荐效果。展开更多
基金co-supported by the Aeronautical Science Foundation of China(Nos.2018ZA52002 and 2019ZA052011)。
文摘Multi-scale problems in Computational Fluid Dynamics(CFD)often require numerous simulations across various design parameters.Using a fixed mesh for all cases may fail to capture critical physical features.Moving mesh adaptation provides an optimal resource allocation to obtain high-resolution flow-fields on low-resolution meshes.However,most existing methods require manual experience and the flow posteriori information poses great challenges to practical applications.In addition,generating adaptive meshes directly from design parameters is difficult due to highly nonlinear relationships.The diffusion model is currently the most popular model in generative tasks that integrates the diffusion principle into deep learning to capture the complex nonlinear correlations.A dual diffusion framework,Para2Mesh,is proposed to predict the adaptive meshes from design parameters by exploiting the robust data distribution learning ability of the diffusion model.Through iterative denoising,the proposed dual networks accurately reconstruct the flow-field to provide flow features as supervised information,and then achieve rapid and reliable mesh movement.Experiments in CFD scenarios demonstrate that Para2Mesh predicts similar meshes directly from design parameters with much higher efficiency than traditional method.It could become a real-time adaptation tool to assist engineering design and optimization,providing a promising solution for high-resolution flow-field analysis.
文摘The formamide-titanium oxide interaction mechanism is a research target of great importance for understanding the elementary events of the origin of life: the synthesis of nucleoside bases and formation of biological molecules needed for life. Titanium oxide (TiO2) can act as a strongly adsorbing surface or a catalytic material. In the present study, a comparative molecular dynamics analysis performed to clarify the adsorbing and diffusion properties of liquid formamide on a TiO2 surface in the presence of water molecules. The structural features of the formamide concentration effect (the accumulation of molecules) on a TiO2 surface in the presence and absence of water solvent are cleared up. Modification of the formamide diffusion abilities mediated by a water solvent is observed to correlate with the formamide-water concentration distribution on the surface.
基金supported in part by the National Natural Science Foundation of China under Grant Nos.62472202 and 61972185the Yunnan Ten Thousand Talents Plan for Young Professionals.
文摘Accurate identification of mutation-driven cancer driver genes is vital for cancer biology and targeted therapy.To address noise in protein–protein interaction(PPI)networks,we propose DGCL_DWA,a novel method utilizing graph diffusion and contrastive learning.DGCL_DWA first employs personalized PageRank to generate a diffusion graph,revealing hidden biological connections.Chebyshev graph convolution extracts features from both the PPI and diffusion networks,and neighborhood contrastive learning harmonizes gene representations,reducing noise.The network-specific features are refined via Chebyshev graph convolutions,which are constrained via node classification and link prediction.A dynamic weight adjustment strategy balances task-specific losses during training.Finally,logistic regression is used to predict driver genes.The experimental results demonstrate the superior performance in pan-cancer and specific cancer driver gene identification compared with state-of-the-art methods.Ablation studies confirm the positive impact of the diffusion graph,contrastive learning,and dynamic weight adjustment on predictive accuracy.The source codes are available at https://doi.org/10.57760/sciencedb.31933.
基金supported in part by the National Natural Science Foundation of China(Grant Nos.72371244,72231011,72301286,72431011 and 72421002).
文摘Locating the source of diffusion in complex networks is a critical and challenging problem,exemplified by tasks such as identifying the origin of power grid faults or detecting the source of computer viruses.The accuracy of source localization in most existing methods is highly dependent on the number of infected nodes.When there are few infected nodes in the network,the accuracy is relatively limited.This poses a major challenge in identifying the source in the early stages of diffusion.This article presents a novel deep learning-based model for source localization under limited information conditions,denoted as GCN-MSL (Graph Convolutional Networks and network Monitor-based Source Localization model).The GCN-MSL model is less affected by the number of infected nodes and enables the efficient identification of the diffusion source in the early stages.First,pre-deployed monitor nodes,controlled by the network administrator,continuously report real-time data,including node states and the arrival time of anomalous signals.These data,along with the network topology,are used to construct node features.Graph convolutional networks are employed to aggregate information from multiple-order neighbors,thereby forming comprehensive node representations.Subsequently,the model is trained with the true source labeled as the target,allowing it to distinguish the source node from other nodes within the network.Once trained,the model can be applied to locate hidden sources in other diffusion networks.Experimental results across multiple data sets demonstrate the superiority of the GCN-MSL model,especially in the early stages of diffusion,where it significantly enhances both the accuracy and efficiency of source localization.Additionally,the GCN-MSL model exhibits strong robustness and adaptability to variations in external parameters of monitor nodes.The proposed method holds significant value in the timely detection of anomalous signals within complex networks and preventing the spread of harmful information.
文摘Renewable energy production and the balance between production and demand have become increasingly crucial in modern power systems,necessitating accurate forecasting.Traditional deterministic methods fail to capture the inherent uncertainties associated with intermittent renewable sources and fluctuating demand patterns.This paper proposes a novel denoising diffusion method for multivariate time series probabilistic forecasting that explicitly models the interdependencies between variables through graph modeling.Our framework employs a parallel feature extraction module that simultaneously captures temporal dynamics and spatial correlations,enabling improved forecasting accuracy.Through extensive evaluation on two world real-datasets focused on renewable energy and electricity demand,we demonstrate that our approach achieves state-of-the-art performance in probabilistic energy time series forecasting tasks.By explicitly modeling variable interdependencies and incorporating temporal information,our method provides reliable probabilistic forecasts,crucial for effective decision-making and resource allocation in the energy sector.Extensive experiments validate that our proposed method reduces the Continuous Ranked Probability Score(CRPS)by 2.1%-70.9%,Mean Absolute Error(MAE)by 4.4%-52.2%,and Root Mean Squared Error(RMSE)by 7.9%-53.4%over existing methods on two real-world datasets.
文摘针对基于知识图谱的推荐系统中存在的高阶建模困难与用户特征建模不足的问题,提出基于多跳机制的扩散图谱推荐模型(a diffusion map recommendation model based on multi-hop mechanism,MultiHop-GDN)。该模型通过端到端方法挖掘知识图谱高阶语义信息,涵盖知识图谱构建、特征提取网络构建与多跳扩散模型构建三部分内容。利用用户特征和项目特征构建知识图谱;深入分析用户兴趣、偏好和历史行为等信息,构建用户画像和兴趣模型;提出特征提取网络捕获深层次语义信息,通过本文模型的计算得到预测值。在两个公开数据集的对比实验表明,MultiHop-GDN能够同时实现用户和项目的高阶建模,与其他代表论文的模型相比有良好的推荐效果。
