Model Representation Research Articles

Research on the Coupled Representation of Model and Data Integration Guided by Metadata Sets

Abstract. Aiming at the problems of islands, redundancy and sharing difficulties of spatial and temporal business models and data in geospatial and spatial business systems, this paper proposes an integrated and coupled expression method of models and data guided by metadata sets, based on the research on core metadata of information resources and component-based encapsulation. First of all, this method classifies the model and data features, and constructs the metadata description and expression framework dominated by feature meta-components; then, establishes metadata sets through the expansion of the metadata description and expression framework; then, constructs the correlation between the model and the data through the use of the established metadata sets by means of the component-based encapsulation method, so as to complete the integrated coupling of the model and the data; and finally, realises the integrated coupling expression of the model and the data in the digital pipeline business system. The results show that this method integrates the expression of digital pipeline element models and data, solves the problems of silos, redundancy and conflicts of heterogeneous models and data from multiple sources, ensures the integrity of the attributes of the models and data elements, and promotes the structured expression and application of the models and data.

One novel transfer learning-based CLIP model combined with self-attention mechanism for differentiating the tumor-stroma ratio in pancreatic ductal adenocarcinoma.

To develop a contrastive language-image pretraining (CLIP) model based on transfer learning and combined with self-attention mechanism to predict the tumor-stroma ratio (TSR) in pancreatic ductal adenocarcinoma on preoperative enhanced CT images, in order to understand the biological characteristics of tumors for risk stratification and guiding feature fusion during artificial intelligence-based model representation. This retrospective study collected a total of 207 PDAC patients from three hospitals. TSR assessments were performed on surgical specimens by pathologists and divided into high TSR and low TSR groups. This study developed one novel CLIP-adapter model that integrates the CLIP paradigm with a self-attention mechanism for better utilizing features from multi-phase imaging, thereby enhancing the accuracy and reliability of tumor-stroma ratio predictions. Additionally, clinical variables, traditional radiomics model and deep learning models (ResNet50, ResNet101, ViT_Base_32, ViT_Base_16) were constructed for comparison. The models showed significant efficacy in predicting TSR in PDAC. The performance of the CLIP-adapter model based on multi-phase feature fusion was superior to that based on any single phase (arterial or venous phase). The CLIP-adapter model outperformed traditional radiomics models and deep learning models, with CLIP-adapter_ViT_Base_32 performing the best, achieving the highest AUC (0.978) and accuracy (0.921) in the test set. Kaplan-Meier survival analysis showed longer overall survival in patients with low TSR compared to those with high TSR. The CLIP-adapter model designed in this study provides a safe and accurate method for predicting the TSR in PDAC. The feature fusion module based on multi-modal (image and text) and multi-phase (arterial and venous phase) significantly improves model performance.

LLM4THP: a computing tool to identify tumor homing peptides by molecular and sequence representation of large language model based on two-layer ensemble model strategy

Tumor homing peptides (THPs) have a distinctive capacity to specifically attach to tumor cells, providing a promising approach for targeted cancer treatment and detection. Although THPs have the potential for significant impact, their detection by conventional methods is both time-consuming and expensive. To tackle this issue, we provide LLM4THP, an innovative computational approach that utilizes large language models (LLMs) to quickly and effectively detect THPs. LLM4THP utilizes two protein LLMs, ESM2 and Prot_T5_XL_UniRef50, to encode peptide sequences. This allows for the capture of complex patterns and relationships within the peptide data. In addition, we utilize inherent sequence characteristics such as Amino Acid Composition (AAC), Pseudo Amino Acid Composition (PAAC), Amphiphilic Pseudo Amino Acid Composition (APAAC), and Composition, Transition, and Distribution (CTD) to improve the representation of peptides. The RDKitDescriptors feature representation approach transforms peptide sequences into molecular objects and computes chemical characteristics, resulting in enhanced THP identification. The LLM4THP ensemble strategy incorporates various features into a two-layer learning architecture. The first layer consists of LightGBM, XGBoost, Random Forest, and Extremely Randomized Trees, which generate a set of meta results. The second layer utilizes Logistic Regression to further refine the identification of sequences as either THP or non-THP. LLM4THP exhibits exceptional performance compared to the most advanced methods, showcasing enhancements in accuracy, Matthew’s correlation coefficient, F1 score, area under the curve, and average precision. The source code and dataset can be accessed at the following URL: https://github.com/abcair/LLM4THP.

A New Framework for Integrating DNN-Based Geographic Simulation Models within GISystems

As a crucial spatial decision support tool, Geographic Information Systems (GISystems) are widely used in fields such as digital watersheds, resource management, environmental assessment, and regional governance, with their core strength lying in the integration of geographic simulation models from various disciplines, enabling the analysis of complex geographical phenomena and the resolution of comprehensive spatial problems. With the rapid advancement of artificial intelligence, deep neural network-based geographic simulation models (DNN-GSMs) have increasingly replaced traditional models, offering significant advantages in simulation accuracy and inference speed, and have become indispensable components in GISystems. However, existing integration methods do not adequately account for the specific characteristics of DNN-GSMs, such as their formats and input/output data types. To address this gap, we propose a novel tight integration framework for DNN-GSMs, comprising four key interfaces: the data representation interface, the model representation interface, the data conversion interface, and the model application interface. These interfaces are designed to describe spatial data, the simulation model, the adaptation between spatial data and the model, and the model’s application process within the GISystem, respectively. To validate the proposed method, we construct a spatial morphology simulation model based on CNN-LSTM, integrate it into a GISystem using the proposed interfaces, and conduct a series of predictive experiments on island morphology evolution. The results demonstrate the effectiveness of the proposed integration framework for DNN-GSMs.

Experiencia didáctica para el estudio y valoración de la Arquitectura Vernácula guatemalteca

The didactic experience for the study and valoration of Guatemalan Vernacular Architecture consists of a series of academic activities in progressive phases, which introduce this type of architectura in danger of extinction to 4th grade students of architecture at the Rafael Landivar University, in the course: valoration of architecture and territory. Students are progressively immersed in the topic, from the bibliography to field visits, endingin the scale representation of an architectural model. This experience allows participant to learn about architecture, its compositional elements, its construction technique and materials, all whitin the framework of the appreciation of material and intangible cultural heritage. ILIA: Investigaciones Latinoamericanas en Ingeniería y Arquitectura, No. 01, 2024: 75-78.

A multi-layer multi-configurational time-dependent Hartree approach to lattice models beyond one dimension.

The multi-layer multi-configurational time-dependent Hartree (MCTDH) approach is an efficient method to study quantum dynamics in real and imaginary time. The present work explores its potential to describe quantum fluids. The multi-layer MCTDH approach in second quantization representation is used to study lattice models beyond one dimension at finite temperatures. A scheme to map the lattice sites onto the MCTDH tree representation for multi-dimensional lattice models is proposed. A statistical sampling scheme previously used in MCTDH calculations is adapted to facilitate an efficient description of the thermal ensemble. As example, a two-dimensional hard-core Bose-Hubbard model is studied considering up to 64 × 64 lattice sites. The single particle function basis set size required to obtain converged results is found to not increase with the lattice size. The numerical results properly simulate the finite temperature Berezinskii-Kosterlitz-Thouless phase transition.

Analysis of representation and generalization capabilities of pre-trained audio models in urban environments

In the last decade, urban noise pollution has become a significant environmental concern that can be mitigated with the help of audio detection algorithms for classifying different sources of noise and creating more informative noise maps. In this context, machine learning, particularly transfer learning, is an essential technology that enables accurate analysis of urban noise sources. However, the choice of the pre-trained model used to compute audio embeddings can significantly influence the performance of downstream classification tasks. This paper aims to compare the embeddings of various pre-trained models on different data collection campaigns in the context of the Sons al balcó project and quantify the robustness of audio representations. To achieve this, we develop metrics and statistically test the presence of distribution shifts in learned latent features. To evaluate the quality of the embeddings, we perform both qualitative and quantitative analysis using dimensionality reduction methods and assess the performance on downstream tasks using data from different collection campaigns. Results highlight major differences between general purpose and specific models. Our findings suggest the need for careful consideration during the choice of the pre-trained model to use in audio event detection applications.

Characterizing Wet Season Precipitation in the Central Amazon Using a Mesoscale Convective System Tracking Algorithm

AbstractTo comprehensively characterize convective precipitation in the central Amazon region, we utilize the Python FLEXible object TRacKeR (PyFLEXTRKR) to track mesoscale convective systems (MCSs) observed through satellite measurements and simulated by the Weather Research and Forecasting model at a convection‐permitting resolution. This study spans a 2‐month period during the wet seasons of 2014 and 2015. We observe a strong correlation between the MCS track density and accumulated precipitation in the Amazon basin. Key factors contributing to precipitation, such as MCS properties (number, size, rainfall intensity, and movement), are thoroughly examined. Our analysis reveals that while the overall model produces fewer MCSs with smaller mean sizes compared to observations, it tends to overpredict total precipitation due to excessive rainfall intensity for heavy rainfall events (≥10 mm hr−1). These biases in simulated MCS properties could vary with the constraints on the convective background environment. Moreover, while the wet bias from heavy (convective) rainfall outweighs the dry bias in light (stratiform) rainfall, the latter can be crucial, particularly when MCS cloud cover is significantly underestimated. A case study for 1 April 2014 highlights the influence of environmental conditions on the MCS lifecycle and identifies an unrealistic model representation in both stratiform and convective precipitation features.

Model-based experiments as epistemic evidence in paleoecology

Where ordinary experiments are impossible and observational data scarce and indirect—particularly in paleoecosystems—computational experiments are often our only means to learn about reality. There are good arguments to count such model-based predictions as evidence, testing hypotheses and updating our beliefs about the world. However, the epistemic weight of computational experiments depends on an adequate model representation of the target system, transparency about predictive uncertainty, and the avoidance of confirmation bias. I argue that mechanistic models are particularly suited for paleoecological predictions but that iterative uncertainty analyses should guide their development. Using a Bayesian framework I propose preregistration and blinded analysis as tools to strengthen the epistemic value of computational experiments. Here, a preregistration marks the boundary between exploratory model development, which establishes credence in the model, and predictive model application, which tests hypotheses. As good modeling practice I suggest clarifying epistemic goals at the outset of a project and accordingly choose methods to maximize the epistemic weight of the computational experiment.

Lack of visual experience leads to severe distortions in the hand representation of the body model

This study investigates the impact of vision on the maintenance of hand representation in the implicit body model, particularly focusing on congenitally blind individuals. To address this, we performed a hand landmark localization task on blind individuals who lacked visual experience of their bodies and compared their performance to normally sighted and normally sighted but blindfolded participants. Through measurements of finger lengths, hand width, and shape index, we demonstrate that blind participants exhibit significantly greater distortions in their hand representation compared to sighted and blindfolded controls. Notably, blind individuals displayed a marked overestimation of hand width and an underestimation of finger lengths, particularly in digits D2, D3, and D4. Surprisingly, blind subjects with partial vision displayed more severe distortions than those with no residual vision. Furthermore, our findings reveal that late-blind participants exhibit similar levels of distortion as congenitally blind individuals, suggesting an extended period of susceptibility to the lack of visual input in shaping body representations. The Reverse Distortion (RD) hypothesis provides a plausible explanation for these distortions, suggesting that compensatory mechanisms occur within the body model to counteract the anisotropic cortical representations. Our results support this hypothesis: blind individuals have expanded cortical representations processing tactile information, so this could lead to more pronounced distortions in their hand representation of the body model. This underscores the importance of visual input in modulating body representations. Overall, our study highlights the malleability of body representations and the intricate interplay between sensory inputs and cortical processing in shaping the implicit body model.

An advanced high dimensional model representation approach for internal combustion engine modeling and optimization

This work introduces an efficient data-driven approach for engine performance modeling and optimization. A highly accurate and robust high-dimensional model representation (HDMR) surrogate model is developed based on the dataset generated from a coupled KIVA4-CHEMKIN Ⅱ software. The HDMR model is integrated with a multi-objective optimization algorithm, enabling the automatic identification of optimal combustion strategies. Results suggest that the constructed HDMR model is able to accurately predict the engine performance and exhaust gas emissions with negligible CPU costs. Moreover, the model effectively identifies optimal engine operating conditions, leading to enhanced fuel efficiency.

A partitioning Monte Carlo approach for consensus tasks in crowdsourcing

The planner’s role in crowdsourcing involves determining the time to stop collecting information (i.e., timed decision, TD). Previous studies have modeled the uncertain crowdsourcing environment as Partially Observable Markov Decision Processes (POMDPs) and utilized the value of information (VOI) to balance the utilities and costs of information collection. However, there is still room for optimization in solving the TD problem within single-agent POMDPs. In this paper, we propose a partitioning Monte Carlo approach for consensus tasks based on the option-candidate (OC) model partitioning. We simplify the state representation for the OC model and introduce a multi-agent POMDP representation. We establish a correspondence between the single-agent and multi-agent models using two consistency theorems. We propose three progressively improved partitioning Monte Carlo (PMC) algorithms to solve the TD problem within the multi-agent POMDP. We conducted experiments in synthetic domains and a citizen science project, demonstrating that the proposed algorithms exhibit continuous improvements in runtime advantages and consistently outperform the SOTA single-agent Monte Carlo sampling-based algorithm while providing nearly consistent output results.

Which to select?: Analysis of speaker representation with graph attention networks.

Although the recent state-of-the-art systems show almost perfect performance, analysis of speaker embeddings has been lacking thus far. An in-depth analysis of speaker representation will be performed by looking into which features are selected. To this end, various intermediate representations of the trained model are observed using graph attentive feature aggregation, which includes a graph attention layer and graph pooling layer followed by a readout operation. To do so, the TIMIT dataset, which has comparably restricted conditions (e.g., the region and phoneme) is used after pre-training the model on the VoxCeleb dataset and then freezing the weight parameters. Through extensive experiments, there is a consistent trend in speaker representation in that the models learn to exploit sequence and phoneme information despite no supervision in that direction. The results shed light to help understand speaker embedding, which is yet considered to be a black box.

Creative and accurate method for optimal hardware implementation of neurons and biological cells: Application in FPGA-based implementation of cardiac pacemaker cell

The sinoatrial (SA) node cells play a vital role as the principal pacemaker in mammalian hearts, generating regular and spontaneous action potentials to regulate the heart’s rhythm. Comprehending the intricate activity of the SA node’s operation necessitates a collection of differential formulas that tackle non-linear functions. The study presents a new technique to improve the digital representation of the SA node cell model, offering benefits such as decreased hardware needs, enhanced processing speed and accuracy, and reduced implementation expenses by transforming the original model’s differential equations into a unified trigonometric function. This transformation significantly simplifies the computational complexity by eliminating the need for multipliers, resulting in a streamlined set of mathematical expressions. The digital implementation of this novel method can be efficiently realized using the Coordinate Rotation Digital Computer (CORDIC) algorithm, which circumvents the necessity for cumbersome mathematical operations. To demonstrate the viability of this approach, the proposed model is successfully synthesized and implemented on a Field-Programmable Gate Array (FPGA). The results of the implementation demonstrate a significant rise in the operating frequency, which is approximately 6.14 times greater than that of the original model. Furthermore, there is a notable 45 percent decrease in power usage. The lowered hardware needs make significant scalability possible, thus allowing for the inclusion of approximately 12 times as many SA node cells on a sole FPGA board in comparison to the original design.

UAV path planning based on improved dung beetle algorithm with multiple strategy integration

This research presents a multi-strategy augmented dung beetle algorithm (CDBO) for UAV path planning in intricate 3D environments. Its main objective is to address the path planning challenge by formulating an integrated cost function model and environment representation that aligns the optimisation task with the navigation prerequisites and safety constraints of the UAV. Firstly, the algorithm initialises the particle population based on the Lévy flight principle to improve the species diversity and adequately search the solution space. Subsequently, an exponentially decreasing inertia weighting strategy is introduced to improve the convergence speed of the algorithm. In addition, the algorithm uses an adaptive [Formula: see text]-distribution with perturbed variation of the updated positions so that the algorithm avoids falling into a local optimum. Finally, to enhance the robustness of the system, the algorithm incorporates an elite retention strategy based on the Pareto principle. In this study, rigorous tests using 12 CEC2022 test functions and Wilcoxon rank sum test are conducted as benchmarks for the performance of the algorithm. The results show that the enhanced dung beetle algorithm outperforms the traditional algorithm, highlighting its efficiency and robustness. In addition, the efficacy of the enhanced dung beetle algorithm was validated in a variety of complex 3D environments, affirming its ability to generate optimal paths. This study further highlights the importance of effective algorithms in UAV path planning, addressing the key issue of optimising flight operations in intricate spatial scenarios.

Seismic reliability analysis using Subset Simulation enhanced with an explorative adaptive conditional sampling algorithm

Reliability analysis of structures under earthquake loading represents a significant engineering challenge. This is due to the required and rather numerically involving non-linear dynamic analysis, the large computational burden when targeting small failure probabilities, and the synthetic earthquake model representation that may contain thousands of random variables. Subset Simulation is an efficient reliability analysis technique that can handle the challenge of a high-dimensional space with a reduced number of structural analysis calls compared to crude Monte Carlo Simulation. In this contribution, firstly, we investigate the conditions for which Subset Simulation performs efficiently. Thereafter we propose an enhancement to the existing Subset Simulation schemes that shows significant potentials for enhancing the strategy for the starting of the Markov Chain Monte Carlo simulations whenever a new level is reached in the Subset Simulation. Finally, the information gathered from the simulations is investigated to verify that Subset Simulation provides meaningful results from a physical point of view.

Jump around: Selecting Markov Chain Monte Carlo parameters and diagnostics for improved food web model quality and ecosystem representation

Capturing ecological data variability in food web models is an important step for improving model representation of empirical systems. One approach is to use linear inverse modelling and Markov Chain Monte Carlo (LIM-MCMC) techniques to set up an inverse LIM problem using empirical data constraints, and then sample multiple plausible food webs from the inverse problem using an MCMC algorithm. We describe the set of plausible food webs as an ‘ensemble’ of solutions to the inverse problem sampled with the LIM-MCMC algorithm. The extent of data variability eventually integrated into an ensemble depends on how well the LIM-MCMC algorithm samples the solution space. Algorithm quality can be adjusted via user-defined parameters describing starting points, jump sizes, and number of iterations or food webs produced. However, little information exists on how each LIM-MCMC algorithm parameter affects the degree of empirical data variability introduced into the ensemble. Further, post hoc algorithm quality diagnostics with commonly used trace plots and the coefficient of variation (CoV) rarely address critical aspects of algorithm quality, such as (1) if the returned ensemble successfully targeted the solution space distribution (stationarity), (2) correlation between ensemble solutions (mixing), and (3) if the ensemble contains enough solutions to adequately capture input data variability (sampling efficiency). Therefore, we used several established MCMC convergence diagnostics to (1) quantify how algorithm parameters affect ensemble flow values and if these differences propagate to ecological indicators and (2) evaluate algorithm quality and compare to current evaluation and ecosystem modelling methods. We applied 30 LIM-MCMC algorithm combinations of varying starting points, jump sizes, and number of iterations to solve food web ensembles from a single food web model. We analysed ensembles with Ecological Network Analysis (ENA) to calculate indicators describing system function. Results show that LIM-MCMC algorithm parameters, in particular the jump size, affect ensemble flow values, which propagate to ecological indicators describing different ecosystem function of the same model. Thereafter, comparisons of post hoc diagnostics show that MCMC convergence diagnostics provided more robust estimates of algorithm quality than trace plots and CoV. Together, these findings underpin several novel recommendations to enhance LIM-MCMC algorithm parameter selection and quality assessments applicable to any ecological ensemble network study.

HADT: Image super-resolution restoration using Hybrid Attention-Dense Connected Transformer Networks

Image super-resolution (SR) plays a vital role in vision tasks, in which Transformer-based methods outperform conventional convolutional neural networks. Existing work usually uses residual linking to improve the performance, but this type of linking provides limited information transfer within the block. Also, existing work usually restricts the self-attention computation to a single window to improve feature extraction. This means transformer-based networks can only use feature information within a limited spatial range. To handle the challenge, this paper proposes a novel Hybrid Attention-Dense Connected Transformer Network (HADT) to utilize the potential feature information better. HADT is constructed by stacking an attentional transformer block (ATB), which contains an Effective Dense Transformer Block (EDTB) and a Hybrid Attention Block (HAB). EDTB combines dense connectivity and swin-transformer to enhance feature transfer and improve model representation, and meanwhile, HAB is used for cross-window information interaction and joint modeling of features for better visualization. Based on the experiments, our method is effective on SR tasks with magnification factors of 2, 3, and 4. For example, using the Urban100 dataset in an experiment with an amplification factor of 4 our method has a PSNR value that is 0.15 dB higher than the previous method and reconstructs a more detailed texture.

Health workforce needs in Malawi: analysis of the Thanzi La Onse integrated epidemiological model of care

BackgroundTo make the best use of health resources, it is crucial to understand the healthcare needs of a population—including how needs will evolve and respond to changing epidemiological context and patient behaviour—and how this compares to the capabilities to deliver healthcare with the existing workforce. Existing approaches to planning either rely on using observed healthcare demand from a fixed historical period or using models to estimate healthcare needs within a narrow domain (e.g., a specific disease area or health programme). A new data-grounded modelling method is proposed by which healthcare needs and the capabilities of the healthcare workforce can be compared and analysed under a range of scenarios: in particular, when there is much greater propensity for healthcare seeking.MethodsA model representation of the healthcare workforce, one that formalises how the time of the different cadres is drawn into the provision of units of healthcare, was integrated with an individual-based epidemiological model—the Thanzi La Onse model—that represents mechanistically the development of disease and ill-health and patients’ healthcare seeking behaviour. The model was applied in Malawi using routinely available data and the estimates of the volume of health service delivered were tested against officially recorded data. Model estimates of the “time needed” and “time available” for each cadre were compared under different assumptions for whether vacant (or established) posts are filled and healthcare seeking behaviour.ResultsThe model estimates of volume of each type of service delivered were in good agreement with the available data. The “time needed” for the healthcare workforce greatly exceeded the “time available” (overall by 1.82-fold), especially for pharmacists (6.37-fold) and clinicians (2.83-fold). This discrepancy would be largely mitigated if all vacant posts were filled, but the large discrepancy would remain for pharmacists (2.49-fold). However, if all of those becoming ill did seek care immediately, the “time needed” would increase dramatically and exceed “time supply” (2.11-fold for nurses and midwives, 5.60-fold for clinicians, 9.98-fold for pharmacists) even when there were no vacant positions.ConclusionsThe results suggest that services are being delivered in less time on average than they should be, or that healthcare workers are working more time than contracted, or a combination of the two. Moreover, the analysis shows that the healthcare system could become overwhelmed if patients were more likely to seek care. It is not yet known what the health consequences of such changes would be but this new model provides—for the first time—a means to examine such questions.

Autoencoder-based inverse design and surrogate-based optimization of an integrated wet granulation manufacturing process

In pharmaceutical manufacturing, integrating model-based design and optimization can be beneficial for accelerating process development. This study explores the utilization of Machine Learning (ML) techniques as a surrogate model for the optimization of a three-unit wet-granulation based flowsheet model for solid dosage form manufacturing. First, a reduced representation of a wet granulation flowsheet model is developed, incorporating a granulation and milling process, along with a novel dissolution model that accounts for the effect of particle size, porosity, and microstructure on dissolution rate. Two optimization approaches are compared, including an autoencoder-based inverse design and a surrogate-based forward optimization. Both methods address the bi-objective problem of maximizing dissolution time and product yield by identifying the optimal granulation and mill process parameters. For this case study, both approaches were effective and incurred a similar computational cost, averaging under 4 s. However, the autoencoder approach offers an advantage through dimensionality reduction, a feature not available in surrogate-based optimization. Dimensional reduction is particularly beneficial for complex process designs with numerous inputs and outputs. The lower dimensional representation helps improve process understanding through enhanced visualization of the process design space and facilitates feasibility studies involving multiple constraints. The autoencoder-based inverse design introduced in this work showcases an implementation of AI and ML in pharmaceutical process development, demonstrating the potential to enhance process efficiency and product quality in complex manufacturing scenarios.