Model Complexity Research Articles

A Review of Smart Camera Sensor Placement in Construction

Cameras, with their low cost and efficiency, are widely used in construction management and structural health monitoring. However, existing reviews on camera sensor placement (CSP) are outdated due to rapid technological advancements. Furthermore, the construction industry poses unique challenges for CSP implementation due to its scale, complexity, and dynamic nature. Previous reviews have not specifically addressed these industry-specific demands. This study aims to fill this gap by analyzing articles from the Web of Science and ASCE databases that focus exclusively on CSP in construction. A rigorous selection process ensures the relevance and quality of the included studies. This comprehensive review navigates through the complexities of camera and environment models, advocating for advanced optimization techniques like genetic algorithms, greedy algorithms, Swarm Intelligence, and Markov Chain Monte Carlo to refine CSP strategies. Simultaneously, Building Information Modeling is employed to consider the progress of construction and visualize optimized layouts, improving the effect of CSP. This paper delves into perspective distortion, the field of view considerations, and the occlusion impacts, proposing a unified framework that bridges practical execution with the theory of optimal CSP. Furthermore, the roadmap for future exploration in the CSP of construction is proposed. This work enriches the study of construction CSP, charting a course for future inquiry, and emphasizes the need for adaptable and technologically congruent CSP approaches amid evolving application landscapes.

Vehicle identification and analysis based on lightweight yolov5 on edge computing platform

Abstract With the accelerating process of urbanization, the role of intelligent transportation systems in urban traffic management has become more crucial. Nevertheless, traditional surveillance cameras only possess the ability to capture videos and lack the capacity for vehicle detection. Consequently, in order to enhance the intelligent capabilities of front-end surveillance cameras and achieve lightweight model deployment, this research has designed a lightweight detection model that can be deployed on embedded devices. Deploying detection models on embedded devices requires a high level of storage and system computing power, and there exist issues such as model speed, model accuracy, model size, and energy efficiency that are difficult to address in existing studies.Specifically, based on YOLOv5, we first use lightweight structures to replace the C3 module and introduce GSConv and Slim Neck structures to replace the Neck of YOLOv5 in order to reduce the model's parameters and computational complexity. After replacing the model structure, we use global channel pruning to remove redundant information, and use asymmetric quantization to convert model parameters from floating point numbers to fixed point numbers, significantly reducing the model's size and load time, and improving the inference speed. For the target tracking part, we combine the ByteTrack target tracking algorithm with the proposed model, effectively reducing the impact of target occlusion on detection and improving the correlation between video frames. In terms of data recording, detection information such as vehicle color, license plate number, and road blockage level can be uploaded to the database server, and the model's detection results can be encoded using the H.264 video encoding format and uploaded to the streaming media server through streaming media protocols. In the experimental part, we selected the embedded Rk3399pro as the deployment object and achieved 97.8% accuracy in the vehicle counting task, meeting the real-time performance requirements of 31FPS.

Agent-based modelling of Mycobacterium tuberculosis transmission: a systematic review

BackgroundTraditional epidemiological models tend to oversimplify the transmission dynamics of Mycobacterium tuberculosis (M.tb) to replicate observed tuberculosis (TB) epidemic patterns. This has led to growing interest in advanced methodologies like agent-based modelling (ABM), which can more accurately represent the complex heterogeneity of TB transmission.ObjectivesTo better understand the use of agent-based models (ABMs) in this context, we conducted a systematic review with two main objectives: (1) to examine how ABMs have been employed to model the intricate heterogeneity of M.tb transmission, and (2) to identify the challenges and opportunities associated with implementing ABMs for M.tb.Search methodsWe conducted a systematic search following PRISMA guidelines across four databases (MEDLINE, EMBASE, Global Health, and Scopus), limiting our review to peer-reviewed articles published in English up to December 2022. Data were extracted by two investigators using a standardized extraction tool. Prospero registration: CRD42022380580.Selection criteriaOur review included peer-reviewed articles in English that implement agent-based, individual-based, or microsimulation models of M.tb transmission. Models focusing solely on in-vitro or within-host dynamics were excluded. Data extraction targeted the methodological, epidemiological, and computational characteristics of ABMs used for TB transmission. A risk of bias assessment was not conducted as the review synthesized modelling studies without pooling epidemiological data.ResultsOur search initially identified 5,077 studies, from which we ultimately included 26 in our final review after exclusions. These studies varied in population settings, time horizons, and model complexity. While many incorporated population heterogeneity and household structures, some lacked essential features like spatial structures or economic evaluations. Only eight studies provided publicly accessible code, highlighting the need for improved transparency.Authors’ conclusionsABMs are a versatile approach for representing complex disease dynamics, particularly in cases like TB, where they address heterogeneous mixing and household transmission often overlooked by traditional models. However, their advanced capabilities come with challenges, including those arising from their stochastic nature, such as parameter tuning and high computational expense. To improve transparency and reproducibility, open-source code sharing, and standardised reporting are recommended to enhance ABM reliability in studying epidemiologically complex diseases like TB.

Robust Model-Free Identification of the Causal Networks Underlying Complex Nonlinear Systems

Inferring causal networks from noisy observations is of vital importance in various fields. Due to the complexity of system modeling, the way in which universal and feasible inference algorithms are studied is a key challenge for network reconstruction. In this study, without any assumptions, we develop a novel model-free framework to uncover only the direct relationships in networked systems from observations of their nonlinear dynamics. Our proposed methods are termed multiple-order Polynomial Conditional Granger Causality (PCGC) and sparse PCGC (SPCGC). PCGC mainly adopts polynomial functions to approximate the whole system model, which can be used to judge the interactions among nodes through subsequent nonlinear Granger causality analysis. For SPCGC, Lasso optimization is first used for dimension reduction, and then PCGC is executed to obtain the final network. Specifically, the conditional variables are fused in this general, model-free framework regardless of their formulations in the system model, which could effectively reconcile the inference of direct interactions with an indirect influence. Based on many classical dynamical systems, the performances of PCGC and SPCGC are analyzed and verified. Generally, the proposed framework could be quite promising for the provision of certain guidance for data-driven modeling with an unknown model.

Explainable AI: Enhancing Interpretability of Machine Learning Models

Explainable Artificial Intelligence (XAI) is emerging as a critical field to address the “black box” nature of many machine learning (ML) models. While these models achieve high predictive accuracy, their opacity undermines trust, adoption, and ethical compliance in critical domains such as healthcare, finance, and autonomous systems. This research explores methodologies and frameworks to enhance the interpretability of ML models, focusing on techniques like feature attribution, surrogate models, and counterfactual explanations. By balancing model complexity and transparency, this study highlights strategies to bridge the gap between performance and explainability. The integration of XAI into ML workflows not only fosters trust but also aligns with regulatory requirements, enabling actionable insights for stakeholders. The findings reveal a roadmap to design inherently interpretable models and tools for post-hoc analysis, offering a sustainable approach to democratize AI.

LLM Supply Chain Provenance: A Blockchain-based Approach

The burgeoning size and complexity of Large Language Models (LLMs) introduce significant challenges in ensuring data integrity. The proliferation of "deep fakes" and manipulated information raises concerns about the vulnerability of LLMs to misinformation. Traditional LLM architectures often lack robust mechanisms for tracking the origin and history of training data. This opaqueness can leave LLMs susceptible to manipulation by malicious actors who inject biased or inaccurate data. This research proposes a novel approach integrating Blockchain Technology (BCT) within the LLM data supply chain. With its core principle of a distributed and immutable ledger, BCT offers a compelling solution to address this challenge. By storing the LLM's data supply chain on a blockchain, we establish a verifiable record of data provenance. This allows for tracing the origin of each data point used to train the LLM, fostering greater transparency and trust in the model's outputs. This decentralised approach minimises the risk of single points of failure and manipulation. Additionally, the immutability of blockchain records ensures that the data provenance remains tamper-proof, further enhancing the trustworthiness of the LLM. Our approach leverages three critical features of BCT to strengthen LLM security: 1) Transaction Anonymity: While data provenance is recorded on the blockchain, identities of data contributors can be anonymised, protecting their privacy while ensuring data integrity. 2) Decentralised Repository: Enhances the system's resilience against potential attacks by distributing the data provenance record across the blockchain network. 3) Block Validation: Rigorous consensus mechanisms ensure the validity of each data point added to the LLM's data supply chain - minimising the risk of incorporating inaccurate or manipulated data into the training process. Using the experimental approach, initial evaluations using simulated LLM training data on a blockchain platform demonstrate the feasibility and effectiveness of the proposed approach in enhancing data integrity. This approach has far-reaching implications for ensuring the trustworthiness of LLMs in various applications.

Semi-supervised learning network for deep-sea nodule mineral image segmentation

The accurate segmentation of deep-sea nodule mineral images is crucial for scientific mining. However, due to the low contrast of deep-sea images and the varying sizes of nodule minerals, existing methods are not effective in segmenting these images. Furthermore, fully supervised deep learning methods require numerous labelled images for training, and labelling deep-sea nodule mineral images is highly difficult, resulting in a scarcity of available labeled images, which limits the model generalization ability. To address these challenges, a semi-supervised learning network for deep-sea nodule image segmentation (NmiNet) was proposed. In this method, a semi-supervised training paradigm based on underwater image enhancement perturbation and uncertainty weighted optimization (UEUO) was designed. This paradigm enabled the model to fully mine the features in many unlabeled nodule mineral images under the condition of few labelled nodule mineral images, improving the model generalization ability. Moreover, a lightweight global and local feature extraction (GLFE) module was designed to enhance the attention of the module to small nodules, and its ability to locate nodules of different scales by fusing local and global features without considerably increasing model complexity. Experimental results on deep-sea nodule mineral images reveal that this method outperforms existing approaches.

Comparison of different machine learning methods in river streamflow estimation using isovel contours and hydraulic variables

ABSTRACT This study evaluates several machine learning (ML) models for estimating streamflow in the Osage River, Missouri, USA, and the Severn River, UK, using hydraulic input variables. These input variables are the flow section area (A), wetted perimeter (Pw ), water surface width (W), and velocity parameter (U) derived from isovel contours. Before using these models, the influential variables are selected using the sequential feature selection (SFS) method to reduce model complexity while maintaining performance. The two most important input variables identified were A and U, reducing the number of inputs from four to two. The results showed that the river’s flow conditions affect the accuracy of ML models used for estimating streamflow. Linear models such as MLR and ANFIS perform better in steady flow conditions (Osage River). In contrast, decision tree-based and non-linear models such as SVR with a radial basis kernel function (SVR_RBF) are better for unsteady flow conditions (Severn River). The finding suggests simpler models outperform complex deep-learning approaches (such as LSTM) for estimating streamflow by hydraulic variables. By selecting appropriate and efficient ML models, hydrometer stations can significantly improve the accuracy of streamflow estimation, leading to more informed decision-making in water management.

A 1T1M-based efficient probability storage and computing cell for the VAE-SBN hybrid model

Abstract Bayesian network are crucial components for uncertainty reasoning and knowledge representation in probabilistic graphical models. As model complexity increases, traditional digital methods face challenges in efficiency and resource utilization. In this work, a novel Bayesian network optimization method based on one-transistor one-memristor (1T1M) units for storing and accessing probability values was proposed. This method leverages the sigmoid-like characteristics of 1T1M for computational optimization, integrates probability storage into sigmoid belief network (SBN), and extends to the variational autoencoder (VAE) framework. By storing and manipulating probabilities at hardware level, this approach enhances storage efficiency and access speed while enhancing computational precision and model expressiveness through the sigmoid-like space of 1T1M. Validation experiments using the MNIST digit generation task demonstrate that this method exhibits unique advantages in SBN structures, particularly in handling complex conditional probabilities and inference tasks. Compared to traditional digital implementations, the proposed method exhibits potential advantages in inference speed and hardware utilization.

Neural Radiance Fields with Hash-Low-Rank Decomposition

In recent advancements in novel view synthesis and neural rendering, neural radiance field (NeRF) has emerged as a powerful technique for synthesizing high-quality novel views of complex 3D scenes. However, the computational and storage demands of NeRF limit its applicability. In this paper, we present a novel approach to NeRF by combining low-rank decomposition and multi-hash encoding through a novel integration process to enhance efficiency and scalability. Our method reduces the model complexity and accelerates the training processes while maintaining high rendering quality. We demonstrate the effectiveness of our approach through extensive experiments on various datasets, showing significant improvements in performance and memory usage compared to traditional NeRF implementations. These results suggest that our approach can make NeRF more practical for real-world applications, such as virtual reality, gaming, and 3D reconstruction.

Cytokine profiles as predictors of HIV incidence using machine learning survival models and statistical interpretable techniques

HIV remains a critical global health issue, with an estimated 39.9 million people living with the virus worldwide by the end of 2023 (according to WHO). Although the epidemic’s impact varies significantly across regions, Africa remains the most affected. In the past decade, considerable efforts have focused on developing preventive measures, such as vaccines and pre-exposure prophylaxis, to combat sexually transmitted HIV. Recently, cytokine profiles have gained attention as potential predictors of HIV incidence due to their involvement in immune regulation and inflammation, presenting new opportunities to enhance preventative strategies. However, the high-dimensional, time-varying nature of cytokine data collected in clinical research, presents challenges for traditional statistical methods like the Cox proportional hazards (PH) model to effectively analyze survival data related to HIV. Machine learning (ML) survival models offer a robust alternative, especially for addressing the limitations of the PH model’s assumptions. In this study, we applied survival support vector machine (SSVM) and random survival forest (RSF) models using changes or means in cytokine levels as predictors to assess their association with HIV incidence, evaluate variable importance, measure predictive accuracy using the concordance index (C-index) and integrated Brier score (IBS) and interpret the model’s predictions using Shapley additive explanations (SHAP) values. Our results indicated that RSFs models outperformed SSVMs models, with the difference covariate model performing better than the mean covariate model. The highest C-index for SSVM was 0.7180 under the difference covariate model, while for RSF, it reached 0.8801 under the difference covariate model using the log-rank split rule. Key cytokines identified as positive predictors of HIV incidence included TNF-A, BASIC-FGF, IL-5, MCP-3, and EOTAXIN, while 29 cytokines were negative predictors. Baseline factors such as condom use frequency, treatment status, number of partners, and sexual activity also emerged as significant predictors. This study underscored the potential of cytokine profiles for predicting HIV incidence and highlighted the advantages of RSFs models in analyzing high-dimensional, time-varying data over SSVMs. It further through ablation studies emphasized the importance of selecting key features within mean and difference based covariate models to achieve an optimal balance between model complexity and predictive accuracy.

Prediction of Higher Education Student Dropout based on Regularized Regression Models

This study explores the critical topic of student dropout in higher education institutions. To allow early and precise interventions and to provide a multifaceted view of student performance, this study combined two predictive models for dropout classification and score prediction. At first, a logistic regression model was developed to predict student dropout at an early stage. Then, to enhance dropout prediction, a second-degree polynomial regression model was used to predict student results based on available academic variables (access, tests, exams, projects, and assignments) from a Moodle course. Dealing with a limited dataset is a key challenge due to the high risk of overfitting. To address this issue and achieve a balance between overfitting, data size, and model complexity, the predictive models were evaluated with L1 (Lasso) and L2 (Ridge) regularization terms. The regularization techniques of the predictive models led to an accuracy of up to 89% and an R2 score of up to 86%.

Issues in federated learning: some experiments and preliminary results

The growing need for data privacy and security in machine learning has led to exploring novel approaches like federated learning (FL) that allow collaborative training on distributed datasets, offering a decentralized alternative to traditional data collection methods. A prime benefit of FL is its emphasis on privacy, enabling data to stay on local devices by moving models instead of data. Despite its pioneering nature, FL faces issues such as diversity in data types, model complexity, privacy concerns, and the need for efficient resource distribution. This paper illustrates an empirical analysis of these challenges within specially designed scenarios, each aimed at studying a specific problem. In particular, differently from existing literature, we isolate the issues that can arise in an FL framework to observe their nature without the interference of external factors.

The Spectrum Difference Enhanced Network for Hyperspectral Anomaly Detection

Most deep learning-based hyperspectral anomaly detection (HAD) methods focus on modeling or reconstructing the hyperspectral background to obtain residual maps from the original hyperspectral images. However, these methods typically do not pay enough attention to the spectral similarity in the complex environment, resulting in inadequate distinction between background and anomalies. Moreover, some anomalies and background are different objects, but they are sometimes recognized as the objects with the same spectrum. To address the issues mentioned above, this paper proposes a Spectrum Difference Enhanced Network (SDENet) for HAD, which employs variational mapping and Transformer to amplify spectrum differences. The proposed network is based on the encoder–decoder structure, which contains a CSWin-Transformer encoder, Variational Mapping Module (VMModule), and CSWin-Transformer decoder. First, the CSWin-Transformer encoder and decoder are designed to supplement image information by extracting deep and semantic features, where a cross-shaped window self-attention mechanism is designed to provide strong modeling capability with minimal computational cost. Second, in order to enhance the spectral difference characteristics between anomalies and background, a randomly sampling VMModule is presented for feature space transformation. Finally, all fully connected mapping operations are replaced with convolutional layers to reduce the model parameters and computational load. The effectiveness of the proposed SDENet is verified on three datasets, and experimental results show that it achieves better detection accuracy and lower model complexity compared with existing methods.

Balancing Complexity, Parsimony, and Applicability in Hydrologic Modeling: A Comparative Evaluation of Four Infiltration Models across Parameterization Scenarios

Balancing Complexity, Parsimony, and Applicability in Hydrologic Modeling: A Comparative Evaluation of Four Infiltration Models across Parameterization Scenarios

Integrating complexity in population modelling: From matrix to dynamic models

Matrix models are widely used in population ecology studies and are valuable for analysing population dynamics, although they are limited in the use of time-varying parameters. This limitation can be overcome by dynamic models. In this study, we revisit a previously published study on a matrix model of a population of the box jellyfish Carybdea marsupialis (L. 1758) in the Western Mediterranean. A dynamic model integrating the transition matrix of the original model is developed in STELLA Architect with the following improvements: (1) Sensitivity study of the reliability of the methodology for calculating the transition matrix and estimation of the errors of the fitting parameters; (2) Closure of the jellyfish life cycle by adding the polyp stage. This will make it possible to simulate scenarios of ecological interest over several years such as a decline in food supply, jellyfish removal strategies, changes in drift currents and changes in substrate availability for planulae to settle. (3) The inclusion of more biological reality. In particular, a temporal pattern of strobilation is added, which improves the fit of the model to the field data.

Selective weighted least square and piecewise bilinear transformation for accurate satellite DSM generation

One of the main products of multi-view stereo (MVS) high-resolution satellite (HRS) images in photogrammetry and remote sensing is digital surface model (DSM). Producing DSMs from MVS HRS images still faces serious challenges due to various reasons such as complexity of imaging geometry and exterior orientation model in HRS, as well as large dimensions and various geometric and illumination variations. The main motivation for conducting this research is to provide a novel and efficient method that enhances the accuracy and completeness of extracting DSM from HRS images compared to existing recent methods. The proposed method called Sat-DSM, consists of five main stages. Initially, a very dense set of tie-points is extracted from the images using a tile-based matching method, phase congruency-based feature detectors and descriptors, and a local geometric consistency correspondence method. Then, the process of Rational Polynomial Coefficients (RPC) block adjustment is performed to compensate the RPC bias errors. After that, a dense matching process is performed to generate 3D point clouds for each pair of input HRS images using a new geometric transformation called PWB (pricewise bilinear) and an accurate area-based matching method called SWLSM (selective weighted least square matching). The key innovations of this research include the introduction of SWLSM and PWB methods for an accurate dense matching process. The PWB is a novel and simple piecewise geometric transformation model based on superpixel over-segmentation that has been proposed for accurate registration of each pair of HRS images. The SWLSM matching method is based on phase congruency measure and a selection strategy to improve the well-known LSM (least square matching) performance. After dense matching process, the final stage is spatial intersection to generate 3D point clouds, followed by elevation interpolation to produce DSM. To evaluate the Sat-DSM method, 12 sets of MVS-HRS data from IRS-P5, ZY3-1, ZY3-2, and Worldview-3 sensors were selected from areas with different landscapes such as urban, mountainous, and agricultural areas. The results indicate the superiority of the proposed Sat-DSM method over four other methods CATALYST, SGM (Semi-global matching), SS-DSM (structural similarity based DSM extraction), and Sat-MVSF in terms of completeness, RMSE, and MEE. The demo code is available at https://www.researchgate.net/publication/377721674_SatDSM.

Markov Progressive Framework, a Universal Paradigm for Modeling Long Videos.

The computational complexity of video models increases linearly with the square number of frames. Thus, constrained bycomputational resources, training video models to learn long-term temporal semantics end-to-end is quite a challenge. Currently, the main-stream method is to split a raw video into clips, leading to incomplete fragmentary temporal information flow and failure of modeling long-term semantics. In this paper, we design the Markov Progressive framework (MaPro), a theoretical framework consisting of the progressive modeling method and a paradigm model tailored for it. Thecore idea of MaPro is to find a paradigm model consisting of proposed Markov operators which can be trained in multiple sequential steps and ensure that the multi-step progressive modeling is equivalent to the conventional end-to-endmodeling. By training the paradigm model under the progressive method, we are able to model long videos end-to-endwith limited resources and ensure the effective transmission of long-term temporal information. We provide implementations of this theoretical system on the mainstream CNN- and Transformer-based models, where they are modified to conform to the Markov paradigm. As a general and robust training method, we experimentally demonstrate that it yields significant performance improvements on different backbones and datasets. As an illustrative example, the proposed method improves the SlowOnly network by 4.1 mAP on Charades and 2.5 top-1 accuracy on Kinetics. And for TimeSformer, MaPro improves its performance on Kinetics by 2.0 top-1 accuracy. Importantly, all these improvements areachieved with a little parameter and computation overhead.

LCFFNet: A Lightweight Cross-scale Feature Fusion Network for human pose estimation

Human pose estimation is one of the most critical and challenging problems in computer vision. It is applied in many computer vision fields and has important research significance. However, it is still a difficult challenge to strike a balance between the number of parameters and computing load of the model and the accuracy of human pose estimation. In this study, we suggest a Lightweight Cross-scale Feature Fusion Network (LCFFNet) to strike a balance between accuracy and computational load and parameter volume.The Lightweight HRNet-Like (LHRNet) network, Cross-Resolution-Aware Semantics Module (CRASM), and Adapt Feature Fusion Module (AFFM) make up LCFFNet. To be more precise, first, we suggest a lightweight LHRNet network that includes Dynamic Multi-scale Convolution Basic (DMSC-Basic block) block, Basic block, and DMSC-Basic block submodules in the network’s three high-resolution subnetwork stages. The proposed dynamic multi-scale convolution in DMSC-Basic block can reduces the amount of model parameters and complexity of the LHRNet network, and has the ability to extract variable pose features. In order to maintain the model’s ability to express features, the Basic block is introduced. As a result, the LHRNet network not only makes the model more lightweight but also enhances its feature expression capabilities. Second, we propose a CRASM module to enhance contextual semantic information while reducing the semantic gap between different scales by fusing features from different scales. Finally, the augmented semantic feature map’s spatial resolution is finally restored from bottom to top using our suggested AFFM, and adaptive feature fusion is used to increase the positioning accuracy of important sites. Our method successfully predicts keypoints with 74.2 % AP, 89.9 % PCKh@0.5 and 66.9 % AP on the MSCOCO 2017, MPII and Crowdpose datasets, respectively. Our model reduces the number of parameters by 89.0 % and the computational complexity by 87.5 % compared with HRNet. The proposed network performs as well as current large-model human pose estimation networks while outperforming state-of the-art lightweight networks.

Abnormality-aware bone fracture detection and classification using the triple context attention model

<span lang="EN-US">In this study, a novel approach is introduced for fracture detection in bone x-ray images, introducing the triple context attention model (TCAN) that combines concentrated extensive convolutional segments with an attention mechanism to enhance positional data. The TCAN model significantly improves fracture recognition accuracy while reducing model complexity. Leveraging a diverse dataset, consistently achieving high accuracy levels across various body parts. By addressing, mislabelling issues, and employing a visual attention network (VAN), to refine the model's performance. The TCAN model emerges as a robust, computationally efficient solution, offering a remarkable average accuracy of 97.86%. This study contributes valuable advancements to medical imaging and diagnostics, providing a highly effective tool for skeletal fracture detection.</span>