Temporal Model Research Articles

Interferometric phase reconstruction with temporal decorrelation model constraints for time-series InSAR surface deformation monitoring

ABSTRACT The time series interferometric synthetic aperture radar (InSAR) (TSInSAR) is an effective technique for monitoring surface deformation with wide coverage and millimeter accuracy. There are two types of scatterers in the SAR field of view, including the permanent scatterer (PS) and the distributed scatterer (DS). Compared to PS with temporarily stable scattering behavior, the filtered interferometric phases are not consistent for DS susceptible to the decorrelation effect. To overcome the phase inconsistency degradation problem of the decorrelated DSs, the phase linking (PL) theory is introduced to recover the equivalent single-reference (ESR) phases from all interferometric combinations. However, the optimization performance of the PL theory is affected by the uncontrollable estimation error of coherence magnitude. Here, we present a novel enhanced PL (EPL) by incorporating with a priori temporal decorrelation model (TDM) to enhance the weight of the short-term interferograms and suppress the abrupt coherence jump with the long temporal baseline. Real experiment selects the Daguangbao landslide triggered by the ‘5.12’ Wenchuan Mw 7.9 earthquake as the study area to verify the algorithm’s effectiveness. Compared to traditional PL methods, the new EPL method can recover the interferometric phase fringe more clearly and obtain a smoother deformation velocity spatially with higher estimation accuracy.

An Improved Transformer Model for Sap Flow Prediction that Efficiently Utilizes Environmental Information

AbstractAs an important reference for assessing plant water consumption and estimating plant transpiration, it is of great significance to achieve accurate prediction of plant sap flow. A number of deep learning models were established and compared using approximately 3 years of continuous eucalyptus flow time series data collected from the SAPFLUXNET open dataset and 6 environmental factors, including shortwave solar incident radiation, air temperature, air relative humidity, net radiation, vapor pressure deficit, and photosynthetic photon flux density. The experimental results show that the improved Transformer model, with the introduction of a two-step self-attention mechanism and simplified design, maintains significant predictive performance advantages compared to the original Transformer model, long short-term memory, gated recurrent unit, and temporal convolutional neural network models. In the shorter 1-h forecast, the mean squared error and coefficient of determination (R2) of the improved Transformer model are 0.0191 and 0.965, respectively. Compared to the suboptimal typical Transformer model, the MSE is reduced by 22.9%, and R2 is increased by 1.0%. Additionally, the improved model maintains stable predictive performance advantages in long-term plant flow prediction. In the longest 8-h advance prediction, the MSE is reduced by 14.9% compared to the suboptimal Transformer model, and R2 increases by 3.0% compared to the Transformer model. The comprehensive experimental results show that the improved Transformer model makes more effective use of environmental information to achieve more accurate and long-term plant flow prediction. This study emphasizes the basic principle and validity of the two-step self-attention network structure and provides a valuable basis for developing more effective methods for predicting plant sap flow.

Crowd Flow Prediction: An Integrated Approach Using Dynamic Spatial–Temporal Adaptive Modeling for Pattern Flow Relationships

ABSTRACTPredicting crowd flows in smart cities poses a significant challenge for the intelligent transportation system (ITS). Traffic management and behavioral analysis are crucial and have garnered considerable attention from researchers. However, accurately and timely predicting crowd flow is difficult due to various complex factors, including dependencies on recent crowd flow and neighboring regions. Existing studies often focus on spatial–temporal dependencies but neglect to model the relationship between crowd flow in distant areas. In our study, we observe that the daily flow of each region remains relatively consistent, and certain regions, despite being far apart, exhibit similar flow patterns, indicating a strong correlation between them. In this paper, we proposed a novel Multiscale Adaptive Graph‐Gated Network (MSAGGN) model. The main components of MSAGGN can be divided into three major parts: (1) To capture the parallel periodic learning architecture through a layer‐wise gated mechanism, a layer‐wise functional approach is employed to modify gated mechanism, establishing parallel skip periodic connections to effectively manage temporal and external factor information at each time interval; (2) a graph convolutional‐based adaptive mechanism that effectively captures crowd flow traffic data by considering dynamic spatial–temporal correlations; and (3) we proposed a novel intelligent channel encoder (ICE). The task of this block is to capture citywide spatial–temporal correlation along external factors to preserve correlation for distant regions with external elements. To integrate spatio‐temporal flexibility, we introduce the adaptive transformation module. We assessed our model's performance by comparing it with previous state‐of‐the‐art models and conducting experiments using two real‐world datasets for evaluation.

Investigation of bubble interaction in a temporal and spatial heat flux partitioning model for subcooled flow boiling

CFD methodology for subcooled flow boiling is significantly affected by the heat flux partitioning model, which plays a crucial role in predicting the mass source term generated by wall boiling of the liquid phase. Although numerous models have been proposed, most research efforts have focused on the spatial dimensions of boiling heat transfer mechanisms, neglecting the temporal aspect. This study presents a heat flux partitioning model for subcooled flow boiling that considers both temporal and spatial dimensions. The model accounts for the contribution of heat flux from the superheated liquid layer during the bubble growth stage, liquid convection during the bubble wait stage within the bubble influence area in the temporal dimension, and heat flux from the overlapping area of bubble influence in the spatial dimension. An approach was developed to determine the bubble influence area by considering the bubble interaction based on the stochastic nature of nucleation sites on the heated wall, verified by the Monte Carlo method. The bubble dynamics parameters were experimentally derived to reduce the uncertainty associated with the boiling sub-models on the calculation. A comparative analysis of boiling curves and wall heat flux partitioning was conducted between the present model and the RPI model under various flow conditions. The results indicate that both models could reasonably predict the boiling curve. However, the RPI model tends to over-predict the proportion of evaporative heat flux, which is considered a weakness. Based on physical principles, the present model accurately captures the fundamental trend of wall heat flux partitioning. This research enhances the understanding of subcooled flow boiling and increases confidence in multiphase CFD methodology predictions.

LSTGCN: Inductive Spatial Temporal Imputation Using Long Short-Term Dependencies

Spatial temporal forecasting of urban sensors is essentially important for many urban systems, such as intelligent transportation and smart cities. However, due to the problem of hardware failure or network failure, there are some missing values or missing monitoring sensors that need to be interpolated. Recent research on deep learning has made substantial progress on imputation problem, especially temporal aspect (i.e., time series imputation), while little attention has been paid to spatial aspect (both dynamic and static) and long-term temporal dependencies. In this article, we proposed a spatial temporal imputation model, named Long Short-Term Graph Convolution Networks (LSTGCN), which includes gated temporal extraction (GTE) module, multi-head attention-based temporal capture (MHAT) module, long-term periodic temporal encoding (LPTE) module, and bidirectional spatial graph convolution (BSGC) module. The GTE adopts a gated mechanism to filter short-term temporal information, while the MHAT utilizes position encoding to enhance the difference of each timestamps, then use multi-head attention to capture short-term temporal dependency. The BSGC is adopted to handle with spatial relationships between sensor nodes. And we design a periodic encoding technique to process long-term temporal dependencies. The BSGC handles spatial relationships between sensor nodes, and a periodic encoding technique is used to process long-term temporal dependencies. Our experimental analysis includes completion and forecasting tasks, as well as transfer and ablation analyses. The results show that our proposed model outperforms state-of-the-art baselines on real-world datasets.

Towards Integrating Federated Learning with Split Learning via Spatio-temporal Graph Framework for Brain Disease Prediction.

Functional Magnetic Resonance Imaging (fMRI) is used for extracting blood oxygen signals from brain regions to map brain functional connectivity for brain disease prediction. Despite its effectiveness, fMRI has not been widely used: on the one hand, collecting and labeling the data is time-consuming and costly, which limits the amount of valid data collected at a single healthcare site; on the other hand, integrating data from multiple sites is challenging due to data privacy restrictions. To address these issues, we propose a novel, integrated Federated learning and Split learning Spatio-temporal Graph framework (FS2G). Specifically, we introduce federated learning and split learning techniques to split a spatio-temporal model into a client temporal model and a server spatial model. In the client temporal model, we propose a time-aware mechanism to focus on changes in brain functional states and use an InceptionTime model to extract information about changes in the brain states of each subject. In the server spatial model, we propose a united graph convolutional network to integrate multiple graph convolutional networks. Integrating federated learning and split learning, FS2G can utilize multi-site fMRI data without violating data privacy protection and reduce the risk of overfitting as it is capable of learning from limited training data sets. Moreover, it boosts the extraction of spatio-temporal features of fMRI using spatio-temporal graph networks. Experiments on ABIDE and ADHD200 datasets demonstrate that our proposed method outperforms state-of-the-art methods. In addition, we explore biomarkers associated with brain disease prediction using community discovery algorithms using intermediate results of FS2G. The source code is available at https://github.com/yutian0315/FS2G.

Supporting Hard Conversations in Close Relationships Through Design

Relationships are perhaps the single greatest source of human happiness, and as part of building strong relationships, conflict and hard conversations are unavoidable. As people increasingly rely on digital communication to initiate and resolve conflicts, we examine how design can improve the experience of working through hard conversations within close relationships. We interviewed six psychotherapists and twenty-one social media users to understand both theoretical best practices for navigating conflict and users' experiences with hard conversations online, particularly on text-based messaging platforms. We used our findings to create a temporal model of how digital design could intervene to support users and their relationships during these conversations. Specifically, we find that design can help to facilitate more mutually consensual difficult conversations, support emotional regulation during the conversation, and help facilitate pauses when necessary. We explore the tensions between balancing the needs of relationships and the individuals in them in digital design, and how to center relationships in digital design.

Reconstruction of Fine-Spatial-Resolution FY-3D-Based Vegetation Indices to Achieve Farmland-Scale Winter Wheat Yield Estimation via Fusion with Sentinel-2 Data

The spatial resolution (250–1000 m) of the FY-3D MERSI is too coarse for agricultural monitoring at the farmland scale (20–30 m). To achieve the winter wheat yield (WWY) at the farmland scale, based on FY-3D, a method framework is developed in this work. The enhanced deep convolutional spatiotemporal fusion network (EDCSTFN) was used to perform a spatiotemporal fusion on the 10 day interval FY-3D and Sentinel-2 vegetation indices (VIs), which were compared with the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM). In addition, a BP neural network was built to calculate the farmland-scale WWY based on the fused VIs, and the Aqua MODIS gross primary productivity product was used as ancillary data for WWY estimation. The results reveal that both the EDCSTFN and ESTARFM achieve satisfactory precision in the fusion of the Sentinel-2 and FY-3D VIs; however, when the period of spatiotemporal data fusion is relatively long, the EDCSTFN can achieve greater precision than ESTARFM. Finally, the WWY estimation results based on the fused VIs show remarkable correlations with the WWY data at the county scale and provide abundant spatial distribution details about the WWY, displaying great potential for accurate farmland-scale WWY estimations based on reconstructed fine-spatial-temporal-resolution FY-3D data.

ConvLSTM-based spatiotemporal and temporal processing models for chaotic vibration prediction of a microbeam

The current work proposes a hybrid data-driven model, consisting of convolutional neural networks (CNN) and convolutional long short-term memory (ConvLSTM), to enable an innovative application in prediction for microbeam's chaotic vibrations. In the proposed CNNConvLSTM model, CNN is used as feature extractor while ConvLSTM retains long-term connectivity across the entire sequence of frames. The flexibility and effectiveness of the model are demonstrated by its capability to perform both spatiotemporal and temporal processing. Specifically, the spatiotemporal model predicts the chaotic vibrations at 19 selected locations within the microbeam at a specific instant, while the temporal model predicts the microbeam's chaotic vibrations at a fixed location over time. Additionally, two conventional spatiotemporal models and three conventional temporal models are built for comparison, demonstrating the superior performance of CNNConvLSTM in terms of shorter training time and lower training and testing loss. The CNNConvLSTM model can provide useful guidance when investigating chaotic vibrations for performance enhancement and design optimization of microbeam-related devices.

On the edge: a multilevel perspective on innovation complexities and dynamic attractors in the supply network

PurposeIn innovation management, the complexity inherited in the supply network may be necessary for success. This study aims to holistically examine innovation complexities and system attractors within a hierarchically nested supply network and explore how they dynamically interact and influence adaptive innovation processes.Design/methodology/approachTaking a complexity theory perspective, we employed a methodological bricolage approach using a single case study with multiple embedded units of analysis – namely, a supply network encompassing 36 firms. We drew upon primary data obtained from 42 interviewees and rich secondary data, and we employed a temporal exponential random graph model to examine the micro-foundations of the evolution of the sampled supply network over a decade.FindingsThis study presents a comprehensive overview of the innovation complexities—relational, temporal, dynamic, operational and structural – and how they manifest within a supply network. It also identifies three systemic attractors – point, periodic and strange – and elucidates their relationships with the complexities and their impact on innovative supply network dynamics. The resulting conceptual framework and working propositions provide a detailed perspective on the complex interplay between balanced order and chaos and the potentially unbalanced innovation states within a supply network.Originality/valueThis research offers an in-depth perspective on the innovation complexities and dynamic attractors within a supply network from a holistic, multilevel perspective. It advances complexity theory and deepens the understanding of supply networks as complex adaptive systems.

Heart Rate Estimation Algorithm Integrating Long and Short-Term Temporal Features

Non-contact heart rate monitoring from facial videos utilizing remote photoplethysmography (rPPG) has gained significant traction in remote health monitoring. Given that rPPG captures the dynamic blood flow within the human body and constitutes a time-series signal characterized by periodic properties, this study introduced a three-dimensional convolutional neural network (3D CNN) designed to simultaneously address long-term periodic and short-term temporal characteristics for effective rPPG signal extraction. Firstly, differential operations are employed to preprocess video data, enhancing the face’s dynamic features. Secondly, building upon the 3D CNN framework, multi-scale dilated convolutions and self-attention mechanisms were integrated to enhance the model’s temporal modeling capabilities further. Finally, interpolation techniques are applied to refine the heart rate calculation methodology. The experiments conducted on the UBFC-rPPG dataset indicate that, compared with the existing optimal algorithm, the average absolute error (MAE) and the root mean square error (RMSE) realized significant enhancements of approximately 28% and 35%. Additionally, through comprehensive analyses such as cross-dataset experiments and complexity analyses, the validity and stability of the proposed algorithm in the task of heart rate estimation were manifested.

L2T-DFM: Learning to Teach with Dynamic Fused Metric

The loss function plays a crucial role in the construction of machine learning algorithms. Employing a teacher model to set loss functions dynamically for student models has attracted attention. In existing works, (1) the characterization of the dynamic loss suffers from some inherent limitations, ie, the computational cost of loss networks and the restricted similarity measurement handcrafted loss functions; and (2) the states of the student model are provided to the teacher model directly without integration, causing the teacher model to underperform when trained on insufficient amounts of data. To alleviate the above-mentioned issues, in this paper, we select and weigh a set of similarity metrics by a confidence-based selection algorithm and a temporal teacher model to enhance the dynamic loss functions. Subsequently, to integrate the states of the student model, we employ statistics to quantify the information loss of the student model. Extensive experiments demonstrate that our approach can enhance student learning and improve the performance of various deep models on real-world tasks, including classification, object detection, and semantic segmentation scenarios.

Temporal Homogenization Modeling of Viscoelastic Asphalt Concretes and Pavement Structures under Large Numbers of Load Cycles

Temporal Homogenization Modeling of Viscoelastic Asphalt Concretes and Pavement Structures under Large Numbers of Load Cycles

Integrating temporal deep learning models for predicting screen-out risk levels in hydraulic fracturing

Amid the transformative shift in global energy structures, the exploitation and utilization of shale gas, an essential unconventional natural gas resource, have drawn widespread attention from both industrial and academic circles. However, screen-out incidents during hydraulic fracturing operations pose significant obstacles to extraction efficiency and safety. Traditional prediction methods, which rely on empirical estimations and simplified models, are deficient in accuracy and real-time applicability. Addressing this, our study introduces a novel deep learning ensemble integrating Gated Recurrent Units (GRU), Transformer, and One-Dimensional Convolutional Neural Networks (1D-CNN) for precise screen-out prediction. This approach markedly improves predictive accuracy by efficiently processing time-series data and capturing the complex dynamics of fracturing processes. Furthermore, the application of the correlation coefficient method and random forest algorithm for feature selection optimizes model input and further enhances prediction accuracy and operational efficiency. Our comparative analysis demonstrates the model’s superiority, achieving an F1 score of 0.951 and a loss of 0.430, clearly surpassing traditional and other deep learning methods. This integration of advanced neural architectures and feature selection techniques not only advances screen-out prediction but also yields practical insights for optimizing shale gas extraction strategies and enhancing safety.

A multiple behaviour temporal network analysis for health behaviours during COVID-19.

The aim of this study was to examine the temporal dynamics of multiple health behaviours (physical activity, alcohol consumption, healthy eating, cigarette consumption, recreational drug use, vaping), and pandemic-related health behaviours (e.g., hand washing, physical distancing) using network psychometrics. The International COVID-19 Awareness and Responses Evaluation (iCARE) study is an international multi-wave observational cohort study of public awareness, attitudes, and responses to public health policies implemented to reduce the spread of COVID-19 on people around the world. A sub-sample of longitudinal data from Canadians (n = 254) was analysed across four waves (February-July 2020). We used temporal network models to fit temporal networks, contemporaneous networks, and between-subject networks from items within the iCARE survey. Positive temporal associations were observed between physical activity and healthy eating, and a bidirectional relationship was evident between outdoor mask use and vaping. A contemporaneous network revealed positive associations between consumption behaviours (vaping, cigarette use, alcohol use, and recreational drug use), and negative associations between physical activity and drug use, and healthy eating and cigarette use. Health behaviours are interconnected and can be modelled as networks or behavioural systems. The application of temporal network analysis to the study of multiple health behaviours is well suited to address key research questions in the field such as 'how do multiple health behaviours co-vary with one another over time'. Future research using time series data and measuring affective and cognitive mediators of behaviour, in addition to health behaviours, has the potential to contribute valuable hypothesis-generating insights.

A novel temporal finite element method to solve static viscoelastic problems

A novel temporal finite element method is presented to solve static viscoelastic problems, which is more flexible and appropriate to describe temporal variation of displacement than conventional finite difference or numerical integral methods. By using virtual work principle, a spatial finite element based governing equation is derived in terms of displacement and its derivatives. Then two temporal finite element models are developed using Gurtin variational principle and weighted residual technique. A kind of hybrid shape functions with polynomial and trigonometric basis is stressed to give more flexible descriptions of time varying variables. A criterion of stability analysis is derived, which can numerically be conducted when the constitution of shape functions and step size are prescribed. A recursive algorithm toward end is developed by which the temporal FE analysis can be conducted via a matrix power product with the initial condition, instead of step marching. The proposed approach is available for the viscoelastic model described by linear differential equations with order h ≤ 2, and can conveniently be combined with well-developed numerical algorithms to tackle with boundary value problems, such as FEM, SBFEM etc. Various numerical examples, including those with static/harmonic loads, creep, stress singularity and heterogeneous structures, etc. are provided to illustrate the efficiency of the proposed approaches, and impacts of the temporal FE model, constitution of shape functions, and step size, etc. are taken into account. Satisfactory results are achieved in comparison with analytical or ABAQUS-based solutions.

ATPAM: Adversarial Temporal Pattern Attention Mechanism

Abstract. Time series forecasting has a large number of applications in daily life, for instance the prediction of stock prices, electricity consumption, exchange rate changes etc. However, the existing time series prediction methods have limitations. The most significant one is that when all the prediction models get the predicted value, a new round of iteration starts after the loss function is calculated with the corresponding real data. This may cause the accumulation of errors, since there is only one loss function to measure the difference between the predicted value and the ground truth, it will make the connection weak and mainly depend on the accuracy of the prediction model. Unfortunately, there are few time series prediction models with high accuracy in reality. To solve the issue, in this paper, we propose a new time series forecasting model Adversarial Temporal Pattern Attention Mechanism (ATPAM), which based on Generative Adversarial Nets (GANs). ATPAM adopts a Temporal Pattern Attention model as the generator to learn time-invariant temporal patterns, and use a discriminator to improve the prediction performance and do auxiliary adversarial training. Extensive experiments on several real-world datasets show the effectiveness of our method.

IGGCN: Individual-Guided Graph Convolution Network for Pedestrian Trajectory Prediction

Accurately predicting the future trajectory of pedestrians is crucial for applications such as autonomous driving and robot navigation. Graph convolution is widely used in trajectory prediction tasks due to its scalability and adaptive feature-learning capabilities. However, there are two problems with pedestrian trajectory prediction methods based on graph convolution: 1. Previous methods struggled to adjust social interactions according to the attributes of different pedestrians, making it difficult to accurately model the relative importance between different pedestrians and others; 2. Previous methods lacked dynamic processing of pedestrian spatial-temporal interaction features to capture high-level spatial-temporal interaction features effectively. Therefore, we propose an Individual-Guided Graph Convolution Network (IGGCN) for pedestrian trajectory prediction. To tackle problem 1, we design an individual-guided interaction module that can adjust pedestrian social interaction modeling according to the pedestrian's attributes, thereby achieving an accurate description of the relative importance of pedestrians. We extend the module to temporal interaction modeling to further achieve an accurate description of the relative importance of time frames. To address problem 2, we design a deformable convolution module to dynamically process spatial-temporal interaction features through deformable convolution kernels, facilitating the capture of high-level spatial-temporal interaction features. We evaluate our method on the ETH, UCY, and SDD datasets. Quantitative analysis shows that our method has lower prediction errors than the current state-of-the-art methods. Qualitative analysis further reveals that our method effectively eliminates the influence of irrelevant pedestrians and accurately models the spatial-temporal interaction relationship of pedestrians.

High spatial–temporal image fusion model for retrieving aerosol optical depth based on top-of-atmosphere reflectance

High spatial–temporal image fusion model for retrieving aerosol optical depth based on top-of-atmosphere reflectance

Temporal Dependency Modeling for Improved Medical Image Segmentation: The R-UNet Perspective

Temporal Dependency Modeling for Improved Medical Image Segmentation: The R-UNet Perspective