Recent engineering applications increasingly adopt smart materials,whose mechanical responses are sensitive to magnetic and electric fields.In this context,new and computationally efficient modeling strategies are ess...Recent engineering applications increasingly adopt smart materials,whose mechanical responses are sensitive to magnetic and electric fields.In this context,new and computationally efficient modeling strategies are essential to predict the multiphysic behavior of advanced structures accurately.Therefore,the manuscript presents a higher-order formulation for the static analysis of laminated anisotropic magneto-electro-elastic doubly-curved shell structures.The fundamental relations account for the full coupling between the electric field,magnetic field,and mechanical elasticity.The configuration variables are expanded along the thickness direction using a generalized formulation based on the Equivalent Layer-Wise approach.Higher-order polynomials are selected,allowing for the assessment of prescribed values of the configuration variables at the top and bottom sides of solids.In addition,an effective strategy is provided for modeling general surface distributions of mechanical pressures and electromagnetic external fluxes.The model is based on a continuum-based formulation which employs an analytical homogenization of the multifield material properties,based on Mori&Tanaka approach,of a magneto-electro-elastic composite material obtained from a piezoelectric and a piezomagnetic phase,with coupled magneto-electro-elastic effects.A semi-analytical Navier solution is applied to the fundamental equations,and an efficient post-processing equilibrium-based procedure is here used,based on the numerical assessment with the Generalized Differential Quadrature(GDQ)method,to recover the response of three-dimensional shells.The formulation is validated through various examples,investigating the multifield response of panels of different curvatures and lamination schemes.An efficient homogenization procedure,based on the Mori&Tanaka approach,is employed to obtain the three-dimensional constitutive relation of magneto-electro-elastic materials.Each model is validated against three-dimensional finite-element simulations,as developed in commercial codes.Furthermore,the full coupling effect between the electric and magnetic response is evaluated via a parametric investigation,with useful insights for design purposes of many engineering applications.The paper,thus,provides a formulation for the magneto-electro-elastic analysis of laminated structures,with a high computational efficiency,since it provides results with three-dimensional capabilities with a two-dimensional formulation.The adoption of higher-order theories,indeed,allows us to efficiently predict not only the mechanical response of the structure as happens in existing literature,but also the through-the-thickness distribution of electric and magnetic variables.A novel higher-order theory has been proposed in this work for the magneto-electro-elastic analysis of laminated shell structures with varying curvatures.This theory employs a generalized method to model the distribution of the displacement field components,electrostatic,and magneto-static potential,accounting for higher-order polynomials.The thickness functions have been defined to prescribe the arbitrary values of configuration variables at the top and bottom surfaces,even though the model is ESL-based.The fundamental governing equations have been derived in curvilinear principal coordinates,considering all coupling effects among different physical phenomena,including piezoelectric,piezomagnetic,and magneto-electric effects.A homogenization algorithm based on a Mori&Tanaka approach has been adopted to obtain the equivalent magneto-electro-mechanical properties of a two-phase transversely isotropic composite.In addition,an effective method has been adopted involving the external loads in terms of surface tractions,as well as the electric and magnetic fluxes.In the post-processing stage,a GDQ-based procedure provides the actual 3D response of a doubly-curved solid.The model has been validated through significant numerical examples,showing that the results of this semi-analytical theory align well with those obtained from 3D numerical models from commercial codes.In particular,the accuracy of the model has been verified for lamination schemes with soft layers and various curvatures under different loading conditions.Moreover,this formulation has been used to predict the effect of combined electric and magnetic loads on the mechanical response of panels with different curvatures and lamination schemes.As a consequence,this theory can be applied in engineering applications where the combined effect of electric and magnetic loads is crucial,thus facilitating their study and design.An existing limitation of this study is that the solution is that it is derived only for structures with uniform curvature,cross-ply lamination scheme,and simply supported boundary conditions.Furthermore,it requires that each lamina within the stacking sequence exhibits magneto-electro-elastic behavior.Therefore,at the present stage,it cannot be used for multifield analysis of classical composite structures with magneto-electric patches.A further enhancement of the research work could be the derivation of a solution employing a numerical technique,to overcome the limitations of the Navier method.In this way,the same theory may be adopted to predict the multifield response of structures with variable curvatures and thickness,as well as anisotropic materials and more complicated boundary conditions.Acknowledgement:The authors are grateful to the Department of Innovation Engineering of Univer-sity of Salento for the support.展开更多
This study presents a generalized two-dimensional model for evaluating the stationary hygro-thermo-mechanical response of laminated shell structures made of advanced materials.It introduces a generalized kinematic mod...This study presents a generalized two-dimensional model for evaluating the stationary hygro-thermo-mechanical response of laminated shell structures made of advanced materials.It introduces a generalized kinematic model,enabling the assessment of arbitrary values of temperature variation and mass concentration variation for the unvaried configuration at the top and bottom surfaces.This is achieved through the Equivalent Layer-Wise description of the unknown field variable using higher-order polynomials and zigzag functions.In addition,an elastic foundation is modeled utilizing the Winkler-Pasternak theory.The fundamental equations,derived from the total free energy of the system,are solved analytically using Navier’s method.Then,the Fourier-based generalized differential quadrature numerical method is adopted to efficiently recover the through-the-thickness distribution of secondary variables in agreement with the hygro-thermal loading conditions.The formulation is applied in some examples of investigation where the response of panels of different curvature and lamination schemes is evaluated under external hygro-thermal fluxes and prescribed values of temperature and moisture concentration.In addition,this study investigates the effect of the hygro-thermal coupling due to Dufour and Soret effect.The present formulation is verified to be a valuable tool for reducing computational effort and determining the effect on the mechanical response of laminated structures in a thermal and hygrometric environment.展开更多
Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interp...Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interpretability.This paper presents an explainable artificial intelligence(XAI)framework that combines a fine-tuned Visual Geometry Group 16-layer network(VGG16)convolutional neural network with layer-wise relevance propagation(LRP)to deliver high-performance classification and transparent decision support.This approach is evaluated on the publicly available Kaggle kidney cancer imaging dataset,which comprises labeled cancerous and noncancerous kidney scans.The proposed model achieved 98.75%overall accuracy,with precision,recall,and F1-score each exceeding 98%on an independent test set.Crucially,LRP-derived heatmaps consistently localize anatomically and pathologically significant regions such as tumor margins in agreement with established clinical criteria.The proposed framework enhances clinician trust by delivering pixel-level justifications alongside state-of-the-art predictive performance.It facilitates informed decision-making,thereby addressing a key barrier to the clinical adoption of DL in oncology.展开更多
Synaptic pruning is a crucial process in synaptic refinement,eliminating unstable synaptic connections in neural circuits.This process is triggered and regulated primarily by spontaneous neural activity and experience...Synaptic pruning is a crucial process in synaptic refinement,eliminating unstable synaptic connections in neural circuits.This process is triggered and regulated primarily by spontaneous neural activity and experience-dependent mechanisms.The pruning process involves multiple molecular signals and a series of regulatory activities governing the“eat me”and“don't eat me”states.Under physiological conditions,the interaction between glial cells and neurons results in the clearance of unnecessary synapses,maintaining normal neural circuit functionality via synaptic pruning.Alterations in genetic and environmental factors can lead to imbalanced synaptic pruning,thus promoting the occurrence and development of autism spectrum disorder,schizophrenia,Alzheimer's disease,and other neurological disorders.In this review,we investigated the molecular mechanisms responsible for synaptic pruning during neural development.We focus on how synaptic pruning can regulate neural circuits and its association with neurological disorders.Furthermore,we discuss the application of emerging optical and imaging technologies to observe synaptic structure and function,as well as their potential for clinical translation.Our aim was to enhance our understanding of synaptic pruning during neural development,including the molecular basis underlying the regulation of synaptic function and the dynamic changes in synaptic density,and to investigate the potential role of these mechanisms in the pathophysiology of neurological diseases,thus providing a theoretical foundation for the treatment of neurological disorders.展开更多
In the presentmanuscript,a Layer-Wise(LW)generalizedmodel is proposed for the linear static analysis of doublycurved shells constrained with general boundary conditions under the influence of concentrated and surface ...In the presentmanuscript,a Layer-Wise(LW)generalizedmodel is proposed for the linear static analysis of doublycurved shells constrained with general boundary conditions under the influence of concentrated and surface loads.The unknown field variable is modelled employing polynomials of various orders,each of them defined within each layer of the structure.As a particular case of the LW model,an Equivalent Single Layer(ESL)formulation is derived too.Different approaches are outlined for the assessment of external forces,as well as for non-conventional constraints.The doubly-curved shell is composed by superimposed generally anisotropic laminae,each of them characterized by an arbitrary orientation.The fundamental governing equations are derived starting from an orthogonal set of principal coordinates.Furthermore,generalized blending functions account for the distortion of the physical domain.The implementation of the fundamental governing equations is performed bymeans of the Generalized Differential Quadrature(GDQ)method,whereas the numerical integrations are computed employing theGeneralized IntegralQuadrature(GIQ)method.In the post-processing phase,an effective procedure is adopted for the reconstruction of stress and strain through-the-thickness distributions based on the exact fulfillment of three-dimensional equilibrium equations.A series of systematic investigations are performed in which the static response of structures with various curvatures and lamination schemes,calculated by the present methodology,have been successfully compared to those ones obtained fromrefined finite element three-dimensional simulations.Even though the present LW approach accounts for a two-dimensional assessment of the structural problem,it is capable of well predicting the three-dimensional response of structures with different characteristics,taking into account a reduced computational cost and pretending to be a valid alternative to widespread numerical implementations.展开更多
Similar to the very vast prior literature on analyzing laminated composite structures,“higher-order”and“layer-wise higher-order”plate and shell theories for functionally-graded(FG)materials and structures are also...Similar to the very vast prior literature on analyzing laminated composite structures,“higher-order”and“layer-wise higher-order”plate and shell theories for functionally-graded(FG)materials and structures are also widely popularized in the literature of the past two decades.However,such higher-order theories involve(1)postulating very complex assumptions for plate/shell kinematics in the thickness direction,(2)defining generalized variables of displacements,strains,and stresses,and(3)developing very complex governing equilibrium,compatibility,and constitutive equations in terms of newly-defined generalized kinematic and generalized kinetic variables.Their industrial applications are thus hindered by their inherent complexity,and the fact that it is difficult for end-users(front-line structural engineers)to completely understand all the newly-defined generalized DOFs for FEM in the higher-order and layer-wise theories.In an entirely different way,very simple 20-node and 27-node 3-D continuum solid-shell elements are developed in this paper,based on the simple theory of 3D solid mechanics,for static and dynamic analyses of functionally-graded plates and shells.A simple Over-Integration(a 4-point Gauss integration in the thickness direction)is used to evaluate the stiffness matrices of each element,while only a single element is used in the thickness direction without increasing the number of degrees of freedom.A stress-recovery approach is used to compute the distribution of transverse stresses by considering the equations of 3D elasticity in Cartesian as well as cylindrical polar coordinates.Comprehensive numerical results are presented for static and dynamic analyses of FG plates and shells,which agree well,either with the existing solutions in the published literature,or with the computationally very expensive solutions obtained by using simple 3D isoparametric elements(with standard Gauss Quadrature)available in NASTRAN(wherein many 3D elements are used in the thickness direction to capture the varying material properties).The effects of the material gradient index,the span-to-thickness ratio,the aspect ratio and the boundary conditions are also studied in the solutions of FG structures.Because the proposed methodology merely involves:(2)standard displacement DOFs at each node,(2)involves a simple 4-point Gaussian over-integration in the thickness direction,(3)relies only on the simple theory of solid mechanics,and(4)is capable of accurately and efficiently predicting the static and dynamical behavior of FG structures in a very simple and cost-effective manner,it is thus believed by the authors that the painstaking and cumbersome development of“higher-order”or“layer-wise higher-order”theories is not entirely necessary for the analyses of FG plates and shells.展开更多
The surge of large-scale models in recent years has led to breakthroughs in numerous fields,but it has also introduced higher computational costs and more complex network architectures.These increasingly large and int...The surge of large-scale models in recent years has led to breakthroughs in numerous fields,but it has also introduced higher computational costs and more complex network architectures.These increasingly large and intricate networks pose challenges for deployment and execution while also exacerbating the issue of network over-parameterization.To address this issue,various network compression techniques have been developed,such as network pruning.A typical pruning algorithm follows a three-step pipeline involving training,pruning,and retraining.Existing methods often directly set the pruned filters to zero during retraining,significantly reducing the parameter space.However,this direct pruning strategy frequently results in irreversible information loss.In the early stages of training,a network still contains much uncertainty,and evaluating filter importance may not be sufficiently rigorous.To manage the pruning process effectively,this paper proposes a flexible neural network pruning algorithm based on the logistic growth differential equation,considering the characteristics of network training.Unlike other pruning algorithms that directly reduce filter weights,this algorithm introduces a three-stage adaptive weight decay strategy inspired by the logistic growth differential equation.It employs a gentle decay rate in the initial training stage,a rapid decay rate during the intermediate stage,and a slower decay rate in the network convergence stage.Additionally,the decay rate is adjusted adaptively based on the filter weights at each stage.By controlling the adaptive decay rate at each stage,the pruning of neural network filters can be effectively managed.In experiments conducted on the CIFAR-10 and ILSVRC-2012 datasets,the pruning of neural networks significantly reduces the floating-point operations while maintaining the same pruning rate.Specifically,when implementing a 30%pruning rate on the ResNet-110 network,the pruned neural network not only decreases floating-point operations by 40.8%but also enhances the classification accuracy by 0.49%compared to the original network.展开更多
3D sparse convolution has emerged as a pivotal technique for efficient voxel-based perception in autonomous systems,enabling selective feature extraction from non-empty voxels while suppressing computational waste.Des...3D sparse convolution has emerged as a pivotal technique for efficient voxel-based perception in autonomous systems,enabling selective feature extraction from non-empty voxels while suppressing computational waste.Despite its theoretical efficiency advantages,practical implementations face under-explored limitations:the fixed geometric patterns of conventional sparse convolutional kernels inevitably process non-contributory positions during sliding-window operations,particularly in regions with uneven point cloud density.To address this,we propose Hierarchical Shape Pruning for 3D Sparse Convolution(HSP-S),which dynamically eliminates redundant kernel stripes through layer-adaptive thresholding.Unlike static soft pruning methods,HSP-S maintains trainable sparsity patterns by progressively adjusting pruning thresholds during optimization,enlarging original parameter search space while removing redundant operations.Extensive experiments validate effectiveness of HSP-S acrossmajor autonomous driving benchmarks.On KITTI’s 3D object detection task,our method reduces 93.47%redundant kernel computations whilemaintaining comparable accuracy(1.56%mAP drop).Remarkably,on themore complexNuScenes benchmark,HSP-S achieves simultaneous computation reduction(21.94%sparsity)and accuracy gains(1.02%mAP(mean Average Precision)and 0.47%NDS(nuScenes detection score)improvement),demonstrating its scalability to diverse perception scenarios.This work establishes the first learnable shape pruning framework that simultaneously enhances computational efficiency and preserves detection accuracy in 3D perception systems.展开更多
The dynamic routing mechanism in evolvable networks enables adaptive reconfiguration of topol-ogical structures and transmission pathways based on real-time task requirements and data character-istics.However,the heig...The dynamic routing mechanism in evolvable networks enables adaptive reconfiguration of topol-ogical structures and transmission pathways based on real-time task requirements and data character-istics.However,the heightened architectural complexity and expanded parameter dimensionality in evolvable networks present significant implementation challenges when deployed in resource-con-strained environments.Due to the critical paths ignored,traditional pruning strategies cannot get a desired trade-off between accuracy and efficiency.For this reason,a critical path retention pruning(CPRP)method is proposed.By deeply traversing the computational graph,the dependency rela-tionship among nodes is derived.Then the nodes are grouped and sorted according to their contribu-tion value.The redundant operations are removed as much as possible while ensuring that the criti-cal path is not affected.As a result,computational efficiency is improved while a higher accuracy is maintained.On the CIFAR benchmark,the experimental results demonstrate that CPRP-induced pruning incurs accuracy degradation below 4.00%,while outperforming traditional feature-agnostic grouping methods by an average 8.98%accuracy improvement.Simultaneously,the pruned model attains a 2.41 times inference acceleration while achieving 48.92%parameter compression and 53.40%floating-point operations(FLOPs)reduction.展开更多
The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classificati...The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classification.However,BERT’s size and computational demands limit its practicality,especially in resource-constrained settings.This research compresses the BERT base model for Bengali emotion classification through knowledge distillation(KD),pruning,and quantization techniques.Despite Bengali being the sixth most spoken language globally,NLP research in this area is limited.Our approach addresses this gap by creating an efficient BERT-based model for Bengali text.We have explored 20 combinations for KD,quantization,and pruning,resulting in improved speedup,fewer parameters,and reduced memory size.Our best results demonstrate significant improvements in both speed and efficiency.For instance,in the case of mBERT,we achieved a 3.87×speedup and 4×compression ratio with a combination of Distil+Prune+Quant that reduced parameters from 178 to 46 M,while the memory size decreased from 711 to 178 MB.These results offer scalable solutions for NLP tasks in various languages and advance the field of model compression,making these models suitable for real-world applications in resource-limited environments.展开更多
Filter pruning effectively compresses the neural network by reducing both its parameters and computational cost.Existing pruning methods typically rely on pre-designed pruning criteria to measure filter importance and...Filter pruning effectively compresses the neural network by reducing both its parameters and computational cost.Existing pruning methods typically rely on pre-designed pruning criteria to measure filter importance and remove those deemed unimportant.However,different layers of the neural network exhibit varying filter distributions,making it inappropriate to implement the same pruning criterion for all layers.Additionally,some approaches apply different criteria from the set of pre-defined pruning rules for different layers,but the limited space leads to the difficulty of covering all layers.If criteria for all layers are manually designed,it is costly and difficult to generalize to other networks.To solve this problem,we present a novel neural network pruning method based on the Criterion Learner and Attention Distillation(CLAD).Specifically,CLAD develops a differentiable criterion learner,which is integrated into each layer of the network.The learner can automatically learn the appropriate pruning criterion according to the filter parameters of each layer,thus the requirement of manual design is eliminated.Furthermore,the criterion learner is trained end-to-end by the gradient optimization algorithm to achieve efficient pruning.In addition,attention distillation,which fully utilizes the knowledge of unpruned networks to guide the optimization of the learner and improve the pruned network performance,is introduced in the process of learner optimization.Experiments conducted on various datasets and networks demonstrate the effectiveness of the proposed method.Notably,CLAD reduces the FLOPs of Res Net-110 by about 53%on the CIFAR-10 dataset,while simultaneously improves the network's accuracy by 0.05%.Moreover,it reduces the FLOPs of Res Net-50 by about 46%on the Image Net-1K dataset,and maintains a top-1 accuracy of 75.45%.展开更多
End-to-end object detection Transformer(DETR)successfully established the paradigm of the Transformer architecture in the field of object detection.Its end-to-end detection process and the idea of set prediction have ...End-to-end object detection Transformer(DETR)successfully established the paradigm of the Transformer architecture in the field of object detection.Its end-to-end detection process and the idea of set prediction have become one of the hottest network architectures in recent years.There has been an abundance of work improving upon DETR.However,DETR and its variants require a substantial amount of memory resources and computational costs,and the vast number of parameters in these networks is unfavorable for model deployment.To address this issue,a greedy pruning(GP)algorithm is proposed,applied to a variant denoising-DETR(DN-DETR),which can eliminate redundant parameters in the Transformer architecture of DN-DETR.Considering the different roles of the multi-head attention(MHA)module and the feed-forward network(FFN)module in the Transformer architecture,a modular greedy pruning(MGP)algorithm is proposed.This algorithm separates the two modules and applies their respective optimal strategies and parameters.The effectiveness of the proposed algorithm is validated on the COCO 2017 dataset.The model obtained through the MGP algorithm reduces the parameters by 49%and the number of floating point operations(FLOPs)by 44%compared to the Transformer architecture of DN-DETR.At the same time,the mean average precision(mAP)of the model increases from 44.1%to 45.3%.展开更多
Accurate traffic flow prediction(TFP)is vital for efficient and sustainable transportation management and the development of intelligent traffic systems.However,missing data in real-world traffic datasets poses a sign...Accurate traffic flow prediction(TFP)is vital for efficient and sustainable transportation management and the development of intelligent traffic systems.However,missing data in real-world traffic datasets poses a significant challenge to maintaining prediction precision.This study introduces REPTF-TMDI,a novel method that combines a Reduced Error Pruning Tree Forest(REPTree Forest)with a newly proposed Time-based Missing Data Imputation(TMDI)approach.The REP Tree Forest,an ensemble learning approach,is tailored for time-related traffic data to enhance predictive accuracy and support the evolution of sustainable urbanmobility solutions.Meanwhile,the TMDI approach exploits temporal patterns to estimate missing values reliably whenever empty fields are encountered.The proposed method was evaluated using hourly traffic flow data from a major U.S.roadway spanning 2012-2018,incorporating temporal features(e.g.,hour,day,month,year,weekday),holiday indicator,and weather conditions(temperature,rain,snow,and cloud coverage).Experimental results demonstrated that the REPTF-TMDI method outperformed conventional imputation techniques across various missing data ratios by achieving an average 11.76%improvement in terms of correlation coefficient(R).Furthermore,REPTree Forest achieved improvements of 68.62%in RMSE and 70.52%in MAE compared to existing state-of-the-art models.These findings highlight the method’s ability to significantly boost traffic flow prediction accuracy,even in the presence of missing data,thereby contributing to the broader objectives of sustainable urban transportation systems.展开更多
基金funded by the Project PNRR M4C2—Innovation Grant DIRECT:Digital twIns foR EmergenCy supporT—CUP F83C22000740001.
文摘Recent engineering applications increasingly adopt smart materials,whose mechanical responses are sensitive to magnetic and electric fields.In this context,new and computationally efficient modeling strategies are essential to predict the multiphysic behavior of advanced structures accurately.Therefore,the manuscript presents a higher-order formulation for the static analysis of laminated anisotropic magneto-electro-elastic doubly-curved shell structures.The fundamental relations account for the full coupling between the electric field,magnetic field,and mechanical elasticity.The configuration variables are expanded along the thickness direction using a generalized formulation based on the Equivalent Layer-Wise approach.Higher-order polynomials are selected,allowing for the assessment of prescribed values of the configuration variables at the top and bottom sides of solids.In addition,an effective strategy is provided for modeling general surface distributions of mechanical pressures and electromagnetic external fluxes.The model is based on a continuum-based formulation which employs an analytical homogenization of the multifield material properties,based on Mori&Tanaka approach,of a magneto-electro-elastic composite material obtained from a piezoelectric and a piezomagnetic phase,with coupled magneto-electro-elastic effects.A semi-analytical Navier solution is applied to the fundamental equations,and an efficient post-processing equilibrium-based procedure is here used,based on the numerical assessment with the Generalized Differential Quadrature(GDQ)method,to recover the response of three-dimensional shells.The formulation is validated through various examples,investigating the multifield response of panels of different curvatures and lamination schemes.An efficient homogenization procedure,based on the Mori&Tanaka approach,is employed to obtain the three-dimensional constitutive relation of magneto-electro-elastic materials.Each model is validated against three-dimensional finite-element simulations,as developed in commercial codes.Furthermore,the full coupling effect between the electric and magnetic response is evaluated via a parametric investigation,with useful insights for design purposes of many engineering applications.The paper,thus,provides a formulation for the magneto-electro-elastic analysis of laminated structures,with a high computational efficiency,since it provides results with three-dimensional capabilities with a two-dimensional formulation.The adoption of higher-order theories,indeed,allows us to efficiently predict not only the mechanical response of the structure as happens in existing literature,but also the through-the-thickness distribution of electric and magnetic variables.A novel higher-order theory has been proposed in this work for the magneto-electro-elastic analysis of laminated shell structures with varying curvatures.This theory employs a generalized method to model the distribution of the displacement field components,electrostatic,and magneto-static potential,accounting for higher-order polynomials.The thickness functions have been defined to prescribe the arbitrary values of configuration variables at the top and bottom surfaces,even though the model is ESL-based.The fundamental governing equations have been derived in curvilinear principal coordinates,considering all coupling effects among different physical phenomena,including piezoelectric,piezomagnetic,and magneto-electric effects.A homogenization algorithm based on a Mori&Tanaka approach has been adopted to obtain the equivalent magneto-electro-mechanical properties of a two-phase transversely isotropic composite.In addition,an effective method has been adopted involving the external loads in terms of surface tractions,as well as the electric and magnetic fluxes.In the post-processing stage,a GDQ-based procedure provides the actual 3D response of a doubly-curved solid.The model has been validated through significant numerical examples,showing that the results of this semi-analytical theory align well with those obtained from 3D numerical models from commercial codes.In particular,the accuracy of the model has been verified for lamination schemes with soft layers and various curvatures under different loading conditions.Moreover,this formulation has been used to predict the effect of combined electric and magnetic loads on the mechanical response of panels with different curvatures and lamination schemes.As a consequence,this theory can be applied in engineering applications where the combined effect of electric and magnetic loads is crucial,thus facilitating their study and design.An existing limitation of this study is that the solution is that it is derived only for structures with uniform curvature,cross-ply lamination scheme,and simply supported boundary conditions.Furthermore,it requires that each lamina within the stacking sequence exhibits magneto-electro-elastic behavior.Therefore,at the present stage,it cannot be used for multifield analysis of classical composite structures with magneto-electric patches.A further enhancement of the research work could be the derivation of a solution employing a numerical technique,to overcome the limitations of the Navier method.In this way,the same theory may be adopted to predict the multifield response of structures with variable curvatures and thickness,as well as anisotropic materials and more complicated boundary conditions.Acknowledgement:The authors are grateful to the Department of Innovation Engineering of Univer-sity of Salento for the support.
基金funded by the Project PNRR M4C2—Innovation grant DIRECT:Digital twIns foR EmergenCy supporT—CUP F83C22000740001.
文摘This study presents a generalized two-dimensional model for evaluating the stationary hygro-thermo-mechanical response of laminated shell structures made of advanced materials.It introduces a generalized kinematic model,enabling the assessment of arbitrary values of temperature variation and mass concentration variation for the unvaried configuration at the top and bottom surfaces.This is achieved through the Equivalent Layer-Wise description of the unknown field variable using higher-order polynomials and zigzag functions.In addition,an elastic foundation is modeled utilizing the Winkler-Pasternak theory.The fundamental equations,derived from the total free energy of the system,are solved analytically using Navier’s method.Then,the Fourier-based generalized differential quadrature numerical method is adopted to efficiently recover the through-the-thickness distribution of secondary variables in agreement with the hygro-thermal loading conditions.The formulation is applied in some examples of investigation where the response of panels of different curvature and lamination schemes is evaluated under external hygro-thermal fluxes and prescribed values of temperature and moisture concentration.In addition,this study investigates the effect of the hygro-thermal coupling due to Dufour and Soret effect.The present formulation is verified to be a valuable tool for reducing computational effort and determining the effect on the mechanical response of laminated structures in a thermal and hygrometric environment.
基金supported through the Ongoing Research Funding Program(ORF-2025-498),King Saud University,Riyadh,Saudi Arabia.
文摘Early and accurate cancer diagnosis through medical imaging is crucial for guiding treatment and enhancing patient survival.However,many state-of-the-art deep learning(DL)methods remain opaque and lack clinical interpretability.This paper presents an explainable artificial intelligence(XAI)framework that combines a fine-tuned Visual Geometry Group 16-layer network(VGG16)convolutional neural network with layer-wise relevance propagation(LRP)to deliver high-performance classification and transparent decision support.This approach is evaluated on the publicly available Kaggle kidney cancer imaging dataset,which comprises labeled cancerous and noncancerous kidney scans.The proposed model achieved 98.75%overall accuracy,with precision,recall,and F1-score each exceeding 98%on an independent test set.Crucially,LRP-derived heatmaps consistently localize anatomically and pathologically significant regions such as tumor margins in agreement with established clinical criteria.The proposed framework enhances clinician trust by delivering pixel-level justifications alongside state-of-the-art predictive performance.It facilitates informed decision-making,thereby addressing a key barrier to the clinical adoption of DL in oncology.
基金supported by the National Natural Science Foundation of China,No.31760290,82160688the Key Development Areas Project of Ganzhou Science and Technology,No.2022B-SF9554(all to XL)。
文摘Synaptic pruning is a crucial process in synaptic refinement,eliminating unstable synaptic connections in neural circuits.This process is triggered and regulated primarily by spontaneous neural activity and experience-dependent mechanisms.The pruning process involves multiple molecular signals and a series of regulatory activities governing the“eat me”and“don't eat me”states.Under physiological conditions,the interaction between glial cells and neurons results in the clearance of unnecessary synapses,maintaining normal neural circuit functionality via synaptic pruning.Alterations in genetic and environmental factors can lead to imbalanced synaptic pruning,thus promoting the occurrence and development of autism spectrum disorder,schizophrenia,Alzheimer's disease,and other neurological disorders.In this review,we investigated the molecular mechanisms responsible for synaptic pruning during neural development.We focus on how synaptic pruning can regulate neural circuits and its association with neurological disorders.Furthermore,we discuss the application of emerging optical and imaging technologies to observe synaptic structure and function,as well as their potential for clinical translation.Our aim was to enhance our understanding of synaptic pruning during neural development,including the molecular basis underlying the regulation of synaptic function and the dynamic changes in synaptic density,and to investigate the potential role of these mechanisms in the pathophysiology of neurological diseases,thus providing a theoretical foundation for the treatment of neurological disorders.
文摘In the presentmanuscript,a Layer-Wise(LW)generalizedmodel is proposed for the linear static analysis of doublycurved shells constrained with general boundary conditions under the influence of concentrated and surface loads.The unknown field variable is modelled employing polynomials of various orders,each of them defined within each layer of the structure.As a particular case of the LW model,an Equivalent Single Layer(ESL)formulation is derived too.Different approaches are outlined for the assessment of external forces,as well as for non-conventional constraints.The doubly-curved shell is composed by superimposed generally anisotropic laminae,each of them characterized by an arbitrary orientation.The fundamental governing equations are derived starting from an orthogonal set of principal coordinates.Furthermore,generalized blending functions account for the distortion of the physical domain.The implementation of the fundamental governing equations is performed bymeans of the Generalized Differential Quadrature(GDQ)method,whereas the numerical integrations are computed employing theGeneralized IntegralQuadrature(GIQ)method.In the post-processing phase,an effective procedure is adopted for the reconstruction of stress and strain through-the-thickness distributions based on the exact fulfillment of three-dimensional equilibrium equations.A series of systematic investigations are performed in which the static response of structures with various curvatures and lamination schemes,calculated by the present methodology,have been successfully compared to those ones obtained fromrefined finite element three-dimensional simulations.Even though the present LW approach accounts for a two-dimensional assessment of the structural problem,it is capable of well predicting the three-dimensional response of structures with different characteristics,taking into account a reduced computational cost and pretending to be a valid alternative to widespread numerical implementations.
基金This research is supported by the Mechanics Section,Vehicle Technology Division,of the US Army Research Labs.The support of National Natural Science Foundation of China(grant No.11502069)Natural Science Foundation of Jiangsu Province(grant No.BK20140838)is also thankfully acknowledged.
文摘Similar to the very vast prior literature on analyzing laminated composite structures,“higher-order”and“layer-wise higher-order”plate and shell theories for functionally-graded(FG)materials and structures are also widely popularized in the literature of the past two decades.However,such higher-order theories involve(1)postulating very complex assumptions for plate/shell kinematics in the thickness direction,(2)defining generalized variables of displacements,strains,and stresses,and(3)developing very complex governing equilibrium,compatibility,and constitutive equations in terms of newly-defined generalized kinematic and generalized kinetic variables.Their industrial applications are thus hindered by their inherent complexity,and the fact that it is difficult for end-users(front-line structural engineers)to completely understand all the newly-defined generalized DOFs for FEM in the higher-order and layer-wise theories.In an entirely different way,very simple 20-node and 27-node 3-D continuum solid-shell elements are developed in this paper,based on the simple theory of 3D solid mechanics,for static and dynamic analyses of functionally-graded plates and shells.A simple Over-Integration(a 4-point Gauss integration in the thickness direction)is used to evaluate the stiffness matrices of each element,while only a single element is used in the thickness direction without increasing the number of degrees of freedom.A stress-recovery approach is used to compute the distribution of transverse stresses by considering the equations of 3D elasticity in Cartesian as well as cylindrical polar coordinates.Comprehensive numerical results are presented for static and dynamic analyses of FG plates and shells,which agree well,either with the existing solutions in the published literature,or with the computationally very expensive solutions obtained by using simple 3D isoparametric elements(with standard Gauss Quadrature)available in NASTRAN(wherein many 3D elements are used in the thickness direction to capture the varying material properties).The effects of the material gradient index,the span-to-thickness ratio,the aspect ratio and the boundary conditions are also studied in the solutions of FG structures.Because the proposed methodology merely involves:(2)standard displacement DOFs at each node,(2)involves a simple 4-point Gaussian over-integration in the thickness direction,(3)relies only on the simple theory of solid mechanics,and(4)is capable of accurately and efficiently predicting the static and dynamical behavior of FG structures in a very simple and cost-effective manner,it is thus believed by the authors that the painstaking and cumbersome development of“higher-order”or“layer-wise higher-order”theories is not entirely necessary for the analyses of FG plates and shells.
基金supported by the National Natural Science Foundation of China under Grant No.62172132.
文摘The surge of large-scale models in recent years has led to breakthroughs in numerous fields,but it has also introduced higher computational costs and more complex network architectures.These increasingly large and intricate networks pose challenges for deployment and execution while also exacerbating the issue of network over-parameterization.To address this issue,various network compression techniques have been developed,such as network pruning.A typical pruning algorithm follows a three-step pipeline involving training,pruning,and retraining.Existing methods often directly set the pruned filters to zero during retraining,significantly reducing the parameter space.However,this direct pruning strategy frequently results in irreversible information loss.In the early stages of training,a network still contains much uncertainty,and evaluating filter importance may not be sufficiently rigorous.To manage the pruning process effectively,this paper proposes a flexible neural network pruning algorithm based on the logistic growth differential equation,considering the characteristics of network training.Unlike other pruning algorithms that directly reduce filter weights,this algorithm introduces a three-stage adaptive weight decay strategy inspired by the logistic growth differential equation.It employs a gentle decay rate in the initial training stage,a rapid decay rate during the intermediate stage,and a slower decay rate in the network convergence stage.Additionally,the decay rate is adjusted adaptively based on the filter weights at each stage.By controlling the adaptive decay rate at each stage,the pruning of neural network filters can be effectively managed.In experiments conducted on the CIFAR-10 and ILSVRC-2012 datasets,the pruning of neural networks significantly reduces the floating-point operations while maintaining the same pruning rate.Specifically,when implementing a 30%pruning rate on the ResNet-110 network,the pruned neural network not only decreases floating-point operations by 40.8%but also enhances the classification accuracy by 0.49%compared to the original network.
文摘3D sparse convolution has emerged as a pivotal technique for efficient voxel-based perception in autonomous systems,enabling selective feature extraction from non-empty voxels while suppressing computational waste.Despite its theoretical efficiency advantages,practical implementations face under-explored limitations:the fixed geometric patterns of conventional sparse convolutional kernels inevitably process non-contributory positions during sliding-window operations,particularly in regions with uneven point cloud density.To address this,we propose Hierarchical Shape Pruning for 3D Sparse Convolution(HSP-S),which dynamically eliminates redundant kernel stripes through layer-adaptive thresholding.Unlike static soft pruning methods,HSP-S maintains trainable sparsity patterns by progressively adjusting pruning thresholds during optimization,enlarging original parameter search space while removing redundant operations.Extensive experiments validate effectiveness of HSP-S acrossmajor autonomous driving benchmarks.On KITTI’s 3D object detection task,our method reduces 93.47%redundant kernel computations whilemaintaining comparable accuracy(1.56%mAP drop).Remarkably,on themore complexNuScenes benchmark,HSP-S achieves simultaneous computation reduction(21.94%sparsity)and accuracy gains(1.02%mAP(mean Average Precision)and 0.47%NDS(nuScenes detection score)improvement),demonstrating its scalability to diverse perception scenarios.This work establishes the first learnable shape pruning framework that simultaneously enhances computational efficiency and preserves detection accuracy in 3D perception systems.
基金Supported by the National Key Research and Development Program of China(No.2022ZD0119003)and the National Natural Science Founda-tion of China(No.61834005).
文摘The dynamic routing mechanism in evolvable networks enables adaptive reconfiguration of topol-ogical structures and transmission pathways based on real-time task requirements and data character-istics.However,the heightened architectural complexity and expanded parameter dimensionality in evolvable networks present significant implementation challenges when deployed in resource-con-strained environments.Due to the critical paths ignored,traditional pruning strategies cannot get a desired trade-off between accuracy and efficiency.For this reason,a critical path retention pruning(CPRP)method is proposed.By deeply traversing the computational graph,the dependency rela-tionship among nodes is derived.Then the nodes are grouped and sorted according to their contribu-tion value.The redundant operations are removed as much as possible while ensuring that the criti-cal path is not affected.As a result,computational efficiency is improved while a higher accuracy is maintained.On the CIFAR benchmark,the experimental results demonstrate that CPRP-induced pruning incurs accuracy degradation below 4.00%,while outperforming traditional feature-agnostic grouping methods by an average 8.98%accuracy improvement.Simultaneously,the pruned model attains a 2.41 times inference acceleration while achieving 48.92%parameter compression and 53.40%floating-point operations(FLOPs)reduction.
文摘The rapid growth of digital data necessitates advanced natural language processing(NLP)models like BERT(Bidi-rectional Encoder Representations from Transformers),known for its superior performance in text classification.However,BERT’s size and computational demands limit its practicality,especially in resource-constrained settings.This research compresses the BERT base model for Bengali emotion classification through knowledge distillation(KD),pruning,and quantization techniques.Despite Bengali being the sixth most spoken language globally,NLP research in this area is limited.Our approach addresses this gap by creating an efficient BERT-based model for Bengali text.We have explored 20 combinations for KD,quantization,and pruning,resulting in improved speedup,fewer parameters,and reduced memory size.Our best results demonstrate significant improvements in both speed and efficiency.For instance,in the case of mBERT,we achieved a 3.87×speedup and 4×compression ratio with a combination of Distil+Prune+Quant that reduced parameters from 178 to 46 M,while the memory size decreased from 711 to 178 MB.These results offer scalable solutions for NLP tasks in various languages and advance the field of model compression,making these models suitable for real-world applications in resource-limited environments.
基金supported in part by the National Natural Science Foundation of China under grants 62073085,61973330 and 62350055in part by the Shenzhen Science and Technology Program,China under grant JCYJ20230807093513027in part by the Fundamental Research Funds for the Central Universities,China under grant 1243300008。
文摘Filter pruning effectively compresses the neural network by reducing both its parameters and computational cost.Existing pruning methods typically rely on pre-designed pruning criteria to measure filter importance and remove those deemed unimportant.However,different layers of the neural network exhibit varying filter distributions,making it inappropriate to implement the same pruning criterion for all layers.Additionally,some approaches apply different criteria from the set of pre-defined pruning rules for different layers,but the limited space leads to the difficulty of covering all layers.If criteria for all layers are manually designed,it is costly and difficult to generalize to other networks.To solve this problem,we present a novel neural network pruning method based on the Criterion Learner and Attention Distillation(CLAD).Specifically,CLAD develops a differentiable criterion learner,which is integrated into each layer of the network.The learner can automatically learn the appropriate pruning criterion according to the filter parameters of each layer,thus the requirement of manual design is eliminated.Furthermore,the criterion learner is trained end-to-end by the gradient optimization algorithm to achieve efficient pruning.In addition,attention distillation,which fully utilizes the knowledge of unpruned networks to guide the optimization of the learner and improve the pruned network performance,is introduced in the process of learner optimization.Experiments conducted on various datasets and networks demonstrate the effectiveness of the proposed method.Notably,CLAD reduces the FLOPs of Res Net-110 by about 53%on the CIFAR-10 dataset,while simultaneously improves the network's accuracy by 0.05%.Moreover,it reduces the FLOPs of Res Net-50 by about 46%on the Image Net-1K dataset,and maintains a top-1 accuracy of 75.45%.
基金Shanghai Municipal Commission of Economy and Information Technology,China(No.202301054)。
文摘End-to-end object detection Transformer(DETR)successfully established the paradigm of the Transformer architecture in the field of object detection.Its end-to-end detection process and the idea of set prediction have become one of the hottest network architectures in recent years.There has been an abundance of work improving upon DETR.However,DETR and its variants require a substantial amount of memory resources and computational costs,and the vast number of parameters in these networks is unfavorable for model deployment.To address this issue,a greedy pruning(GP)algorithm is proposed,applied to a variant denoising-DETR(DN-DETR),which can eliminate redundant parameters in the Transformer architecture of DN-DETR.Considering the different roles of the multi-head attention(MHA)module and the feed-forward network(FFN)module in the Transformer architecture,a modular greedy pruning(MGP)algorithm is proposed.This algorithm separates the two modules and applies their respective optimal strategies and parameters.The effectiveness of the proposed algorithm is validated on the COCO 2017 dataset.The model obtained through the MGP algorithm reduces the parameters by 49%and the number of floating point operations(FLOPs)by 44%compared to the Transformer architecture of DN-DETR.At the same time,the mean average precision(mAP)of the model increases from 44.1%to 45.3%.
文摘Accurate traffic flow prediction(TFP)is vital for efficient and sustainable transportation management and the development of intelligent traffic systems.However,missing data in real-world traffic datasets poses a significant challenge to maintaining prediction precision.This study introduces REPTF-TMDI,a novel method that combines a Reduced Error Pruning Tree Forest(REPTree Forest)with a newly proposed Time-based Missing Data Imputation(TMDI)approach.The REP Tree Forest,an ensemble learning approach,is tailored for time-related traffic data to enhance predictive accuracy and support the evolution of sustainable urbanmobility solutions.Meanwhile,the TMDI approach exploits temporal patterns to estimate missing values reliably whenever empty fields are encountered.The proposed method was evaluated using hourly traffic flow data from a major U.S.roadway spanning 2012-2018,incorporating temporal features(e.g.,hour,day,month,year,weekday),holiday indicator,and weather conditions(temperature,rain,snow,and cloud coverage).Experimental results demonstrated that the REPTF-TMDI method outperformed conventional imputation techniques across various missing data ratios by achieving an average 11.76%improvement in terms of correlation coefficient(R).Furthermore,REPTree Forest achieved improvements of 68.62%in RMSE and 70.52%in MAE compared to existing state-of-the-art models.These findings highlight the method’s ability to significantly boost traffic flow prediction accuracy,even in the presence of missing data,thereby contributing to the broader objectives of sustainable urban transportation systems.
