Computational Resources Research Articles

Optimizing Machine Learning Models with Data-level Approximate Computing: The Role of Diverse Sampling, Precision Scaling, Quantization and Feature Selection Strategies

Efficiency, low-power consumption, and real-time processing in embedded machine learning implementations are critical, particularly for models deployed in environments with large-scale data processing and resource-constrained environments. This paper investigates the application of approximate computing techniques as a viable solution to reduce computational complexity and optimize machine learning models, focusing on two widely used supervised machine learning models: k-nearest neighbors (KNN) and support vector machines (SVM). Although many studies compare machine learning classification techniques, the combined use of optimization strategies remains underexplored. Specifically, the combined utilization of feature selection, sampling, quantization, precision scaling, and relaxation methods for the purpose of optimizing and acquiring training and validation data is underexplored, particularly within the context of medical diagnosis datasets. In this paper, we propose a framework that uses data-level approximate computing techniques, including by diverse sampling strategies, precision scaling, quantization, and feature selection methods, to evaluate the impact of these techniques on the computational efficiency and accuracy of KNN and SVM models. Experimental results demonstrate that with careful application of approximate computing strategies, especially in critical applications such as medical diagnosis, it is possible to achieve considerable gains in efficiency while maintaining acceptable levels of accuracy. The combined application of these methods by selecting 3 features and quantizing the data values to 8 levels, then applying random sampling with 30% reductions and scaling the precision at 5 bits resulted in reductions of 87.5% in computation, 76.9% in memory usage, and 17% in delay, without any degradation in accuracy, as validated by tenfold cross-validation, Training Data validation, and full dataset validation. This study confirms the potential of approximate computing to optimize machine learning workflows, making it particularly suitable for applications with limited computational resources. The source code is publicly available online https://github.com/AyadMDalloo/DatalvlAxC.

Uncertainty analysis of geomechanical responses: China’s first offshore carbon capture and storage project

Enping 15–1 project is China’s first Carbon Capture and Storage (CCS) project to store CO2 in the offshore deep saline aquifer formations. Due to the lack of detailed observed data, geomechanical responses are fraught with uncertainty, which can result in unanticipated risks during the implementation of CO2 injection. Thus, the uncertainty analysis of geomechanical responses shows great importance and necessity for CO2 storage projects. To bridge the gap, in this work, a Thermal-Hydraulic-Mechanical-Chemical (THMC) coupling model was built to study the uncertainty of geomechanical responses including surface uplift and rock damage. First, Monte Carlo sampling technology was used to create 616 cases. Then the sensitivity of 13 parameters (including reservoir, cap, and operating parameters) on geological responses was analysed using the Sobol sensitivity analysis method. The results show that 90.10 % of cases in the uncertainty analysis have a maximum surface uplift of less than 8 mm, and 80.84 % of cases have a safety factor higher than 0.5. However, it is the most likely to cause surface uplift over 100 mm and rock failure when the reservoir permeability is 1 mD. Therefore, the local low permeable zone is at more risk of rock damage and geological hazards. In addition, five parameters (i.e., reservoir permeability, reservoir Young’s modulus, initial horizontal/vertical stress ratio, cap cohesion, and well length) have large contributions to the surface uplift. Three parameters (i.e., Reservoir permeability, reservoir cohesion, and initial horizontal/vertical stress ratio) have considerable effects on the reservoir safety factor. Eventually, proxy models using Regression Trees (RTs) were proposed to predict the surface uplift and reservoir safety factor, which is efficient and saves computational resources. This work provides insights and suggestions into the geomechanical responses of CCS project.

Full 2D optimized window coefficients for improved range velocity estimation in 5G joint communication and sensing

Abstract This work presents an approach for optimization of window coefficients for 5G user equipment side sensing, using orthogonal frequency division multiplexing radar-based range and velocity estimation, based on the sounding reference signal (SRS) from the 5G New Radio (NR) standard. The signal configuration and the corresponding waveform are generated in compliance with the 3rd Generation Partnership Project (3GPP) standard for 5G. The limitations of conventional signal processing for resources available for sensing with the SRS are highlighted. The proposed approach, which optimizes the window coefficients to improve the sensing capabilities, is implemented through two methods. The first method employs a decoupled optimization strategy for range and velocity, showing high computational efficiency. Our results demonstrate that this method significantly improves the peak sidelobe level (PSL) of the velocity profile by over $\mathrm{15}\,\mathrm{dB}$ , although it does not address the issue of diagonally located sidelobes, which occur due to non-uniform resource distribution. The second method adopts a comprehensive full 2D optimization technique. While it requires more computational resources and does not improve the PSL beyond the first method’s achievements, it mitigates the diagonally located sidelobes issue. The level of these have been improved by more than $\mathrm{3}\,\mathrm{dB}$ .

Comparative analysis of AI-based search algorithms in solving 8 puzzle problems

BackgroundThe 8-puzzle problem is a well-known combinatorial search problem, often used to test the effectiveness of various artificial intelligence (AI) algorithms. This study aims to evaluate the performance of three AI-based search algorithms—Breadth-First Search (BFS), Depth-First Search (DFS), and A* Search—in solving the 8-puzzle problem. The algorithms were implemented in Java, and their performance was measured in terms of solution length, the number of expanded nodes, and execution time.ResultsExperiments were conducted on eight randomly generated instances of the 8-puzzle problem. The findings reveal that both BFS and A* Search consistently identify optimal solutions in all cases. However, BFS was found to use more memory and operate slower than A* Search. Conversely, DFS demonstrated faster execution times and lower memory usage compared to BFS but was less reliable in finding optimal or feasible solutions. The study also explored the impact of different heuristic functions on the performance of A* Search. Among the heuristics tested, the Manhattan distance heuristic was determined to be the most effective, offering the best balance between accuracy and efficiency.ConclusionsIn conclusion, both BFS and A* Search are effective in finding optimal solutions, with A* Search being more efficient in terms of speed and memory usage. DFS, while faster and more memory-efficient, is less consistent in producing optimal solutions. The Manhattan distance heuristic significantly enhances the performance of A* Search, making it the preferred heuristic for the 8-puzzle problem. These results suggest that A* Search, particularly with the Manhattan distance heuristic, is a highly efficient choice for solving the 8-puzzle problem, especially in contexts where computational resources are limited.

Additive manufacturing simulations: An approach based on space partitioning and dynamic 3D mesh adaptation

Simulation is one of the most widely used methods for process optimization towards improved part quality in Additive Manufacturing (AM), particularly for metal parts. However, due to the nature of the AM processes and the complex phenomena that occur, simulations that are capable of providing a detailed overview of the physical mechanisms demand considerable computational resources and time. In this study, a numerical approach is presented, which can be applied to any implicit numerical thermal simulation for AM, allowing for a significant decrease in computational time (higher than 70%) with minimal impact on accuracy. This is achieved by combining space partitioning, enabled by a boundary condition that was developed, with dynamic mesh adaptation in the x-, y-, and z-axis. The methodology is described in detail and both the decrease in computational time and the accuracy of the developed approach are validated in a computational case study, as well as using experimental results.

Rapid Attitude Controller Design Enabled by Flight Data

Abstract In this work, we present a new approach that enables rapid learning and design of the attitude controller for quadcopters. The proposed technique leverages the dynamic structure of the system to efficiently learn an accurate linear model around the hovering position from a small batch of flight data. An LQR-based attitude controller is designed based on the estimation accordingly. In our approach, we may not need to conduct multiple experiments to fine-tune the controller and can accomplish such a control design with a single experiment. We further develop this approach into an onboard design framework where the attitude controller can be directly learned online and recursively fine-tunes itself with the limited on board computation resources. The practicality and efficacy of the proposed approaches are demonstrated in both simulation and a physical quadcopter platform.

A novel machine learning workflow to optimize cooling devices grounded in solid-state physics

Cooling devices grounded in solid-state physics are promising candidates for integrated-chip nanocooling applications. These devices are modeled by coupling the quantum non-equilibirum Green’s function for electrons with the heat equation (NEGF+H), which allows to accurately describe the energetic and thermal properties. We propose a novel machine learning (ML) workflow to accelerate the design optimization process of these cooling devices, alleviating the high computational demands of NEGF+H. This methodology, trained with NEGF+H data, obtains the optimum heterostructure designs that provide the best trade-off between the cooling power of the lattice (CP) and the electron temperature (\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$ {\ ext{T}}_{e} $$\\end{document}). Using a vast search space of \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$1.18 \ imes 10^{-5}$$\\end{document} different device configurations, we obtained a set of optimum devices with prediction relative errors lower than \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${4}\\,\\%$$\\end{document} for CP and \\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$${1}\\,\\%$$\\end{document} for Te. The ML workflow reduces the computational resources needed, from two days for a single NEGF+H simulation to 10 s to find the optimum designs.

Active Learning with Particle Swarm Optimization for Enhanced Skin Cancer Classification Utilizing Deep CNN Models.

Skin cancer is a critical global health issue, with millions of non-melanoma and melanoma cases diagnosed annually. Early detection is essential to improving patient outcomes, yet traditional deep learning models for skin cancer classification are often limited by the need for large, annotated datasets and extensive computational resources. The aim of this study is to address these limitations by proposing an efficient skin cancer classification framework that integrates active learning (AL) with particle swarm optimization (PSO). The AL framework selectively identifies the most informative unlabeled instances for expert annotation, minimizing labeling costs while optimizing classifier performance. PSO, a nature-inspired metaheuristic algorithm, enhances the selection process within AL, ensuring the most relevant data points are chosen. This method was applied to train multiple Convolutional Neural Network (CNN) models on the HAM10000 skin lesion dataset. Experimental results demonstrate that the proposed AL-PSO approach significantly improves classification accuracy, with the Least Confidence strategy achieving approximately 89.4% accuracy while using only 40% of the labeled training data. This represents a substantial improvement over traditional approaches in terms of both accuracy and efficiency. The findings indicate that the integration of AL and PSO can accelerate the adoption of AI in clinical settings for skin cancer detection. The code for this study is publicly available at ( https://github.com/Sayantani-31/AL-PSO ).

Application of Machine Learning in Construction Productivity at Activity Level: A Critical Review

There are two crucial resources (i.e., labor and equipment) of productivity in the construction industry. Productivity modeling of these resources would aid stakeholders in project management and improve construction scheduling and monitoring. Hence, this research aims to review machine learning (ML) applications in the process of construction productivity modeling (CPM) for construction labor productivity (CLP) and construction equipment productivity (CEP) from dataset acquisition to data analysis and evaluation, which includes their trends and applicability. An extensive analysis of 131 journals focused on the application of machine learning in construction productivity (ML-CP) from 1990 to 2024 via a mixed review methodology (bibliometric analysis and systematic review) was conducted. It can be concluded that despite the rise in automated dataset collection, the traditional method has its advantages. The review further found that the selection of ML models relies on each particular application, available data, and computational resources. Noticeably, artificial neural networks, convolutional neural networks, support vector machines, and even deep learning demonstrating have been adopted due to their effectiveness in different functionalities and processes in CPM. This study will supplement the insights gained in the review with a comprehensive understanding of how ML applications operate at each stage of CPM, enabling researchers to make future improvements.

A lightweight convolutional neural network for road surface classification under shadow interference

The development of intelligent driving, especially in the intelligent control of active suspension, heavily relies on the predictive perception of upcoming road conditions. To achieve accurate real-time road surface classification and overcome shadow interference, a lightweight convolutional neural network (CNN) based on a novel data augmentation method is proposed and an improved cycle-consistent adversarial network (CycleGAN) is developed to generate shadowed pavement data. The CycleGAN network structure is optimized using the texture self-supervised (TSS) mechanism and the learned perceptual image patch similarity (LPIPS) function, with label smoothing applied during training. The images produced by this data augmentation method closely resemble real-world images. Furthermore, Efficient-MBConv, which offers the advantages of fewer parameters and higher precision, is proposed. Finally, the Light-EfficientNet architecture, based on Efficient-MBConv, is developed and trained on the augmented dataset. Compared with EfficientNet-B0, the number of parameters in Light-EfficientNet is reduced by 61.94 %. The Light-EfficientNet model trained with data augmentation demonstrates an average classification accuracy improvement of 5.76 % on the test set with shadows, compared with the model trained without data augmentation. This approach effectively reduces the impact of shadows on road classification at a lower cost, while also significantly reducing the computational resources required by the CNN, providing real-time and accurate road surface information for the control of active suspension height and damping.

Causal discovery from nonstationary time series

AbstractThis paper introduces a new causal structure learning method for nonstationary time series data, a common data type found in fields such as finance, economics, healthcare, and environmental science. Our work builds upon the constraint-based causal discovery from nonstationary data algorithm (CD-NOD). We introduce a refined version (CD-NOTS) which is designed specifically to account for lagged dependencies in time series data. We compare the performance of different algorithmic choices, such as the type of conditional independence test and the significance level, to help select the best hyperparameters given various scenarios of sample size, problem dimensionality, and availability of computational resources. Using the results from the simulated data, we apply CD-NOTS to a broad range of real-world financial applications in order to identify causal connections among nonstationary time series data, thereby illustrating applications in factor-based investing, portfolio diversification, and comprehension of market dynamics.

Application of quantitative histomorphometric features in computational pathology

AbstractComputer vision has facilitated the execution of various computer‐aided diagnostic tasks. From a methodological perspective, these tasks are primarily implemented using two dominant strategies: end‐to‐end Deep learning (DL)‐based methods and traditional feature engineering‐based methods. DL methods are capable of automatically extracting, analyzing, and filtering features, leading to final decision‐making from whole slide images. However, these methods are often criticized for the “black box” issue, a significant limitation of DL. In contrast, traditional feature engineering‐based methods involve well‐defined quantitative input features. But it was considered as less potent than DL methods. Advances in segmentation technology and the development of quantitative histomorphometric (QH) feature representation have propelled the evolution of feature engineering‐based methods. This review contrasts the performance differences between the two methods and traces the development of QH feature representation. The conclusion is that, with the ongoing progress in QH feature representation and segmentation technology, methods based on QH features will leverage their advantages—such as explainability, reduced reliance on large training datasets, and lower computational resource requirements—to play a more significant role in some clinical tasks. They may even replace DL methods somewhat or be used in conjunction with them to achieve accurate and understandable results.

Topology and size optimization of viscous dampers with discrete design variables by outer approximation

This article proposes a novel seismic retrofitting approach for the topology and size optimization of fluid viscous dampers using discrete design variables. The design variables are the damping coefficients of the dampers, and they are chosen from catalogues, leading to practical optimal design solutions. The optimization problems considered are formulated as 0–1 nonlinear programming problems. An algorithm based on outer approximation and that relies on first-order information is proposed to find optimized designs with modest computational resources and time. The effectiveness of the proposed approach is demonstrated through optimal seismic retrofitting of three-dimensional irregular frame structures. In the optimization problem formulation, the retrofitted structures are subjected to realistic seismic ground acceleration records with constraints on inter-storey drifts and total storey accelerations.

LED-Net: A lightweight edge detection network

As a fundamental task in computer vision, edge detection is becoming increasingly vital in many fields. Recently, large-parameter pre-training models have been used in edge detection tasks. However, significant computational resources are required. This paper presents a Lightweight Edge Detection Network (LED-Net) with only 50K parameters. It mainly consists of three blocks: Coordinate Depthwise Separable Convolution Block (CDSCB), Sample Depthwise Separable Convolution Block (SDSCB), and Feature Fusion Block (FFB). The CDSCB extracts multi-scale features with positional information, thus reducing the time complexity while guaranteeing the performance. Furthermore, SDSCB is adopted to rescale the multi-scale features to a unified resolution efficiently. To obtain refined edge lines, the FFB is adopted to aggregate the features. In addition, a unified loss function is proposed to achieve a thinner edge prediction. By training on the BIPED dataset and evaluating on the UDED dataset, results show that the proposed LED-Net achieves superior performance in both ODS (0.839), OIS (0.855), and AP (0.830).

Reduced-Order Model of Coal Seam Gas Extraction Pressure Distribution Based on Deep Neural Networks and Convolutional Autoencoders

There has been extensive research on the partial differential equations governing the theory of gas flow in coal mines. However, the traditional Proper Orthogonal Decomposition–Radial Basis Function (POD-RBF) reduced-order algorithm requires significant computational resources and is inefficient when calculating high-dimensional data for coal mine gas pressure fields. To achieve the rapid computation of gas extraction pressure fields, this paper proposes a model reduction method based on deep neural networks (DNNs) and convolutional autoencoders (CAEs). The CAE is used to compress and reconstruct full-order numerical solutions for coal mine gas extraction, while the DNN is employed to establish the nonlinear mapping between the physical parameters of gas extraction and the latent space parameters of the reduced-order model. The DNN-CAE model is applied to the reduced-order modeling of gas extraction flow–solid coupling mathematical models in coal mines. A full-order model pressure field numerical dataset for gas extraction was constructed, and optimal hyperparameters for the pressure field reconstruction model and latent space parameter prediction model were determined through hyperparameter testing. The performance of the DNN-CAE model order reduction algorithm was compared to the POD-RBF model order reduction algorithm. The results indicate that the DNN-CAE method has certain advantages over the traditional POD-RBF method in terms of pressure field reconstruction accuracy, overall structure retention, extremum capture, and computational efficiency.

Explainable Pre-Trained Language Models for Sentiment Analysis in Low-Resourced Languages

Sentiment analysis is a crucial tool for measuring public opinion and understanding human communication across digital social media platforms. However, due to linguistic complexities and limited data or computational resources, it is under-represented in many African languages. While state-of-the-art Afrocentric pre-trained language models (PLMs) have been developed for various natural language processing (NLP) tasks, their applications in eXplainable Artificial Intelligence (XAI) remain largely unexplored. In this study, we propose a novel approach that combines Afrocentric PLMs with XAI techniques for sentiment analysis. We demonstrate the effectiveness of incorporating attention mechanisms and visualization techniques in improving the transparency, trustworthiness, and decision-making capabilities of transformer-based models when making sentiment predictions. To validate our approach, we employ the SAfriSenti corpus, a multilingual sentiment dataset for South African under-resourced languages, and perform a series of sentiment analysis experiments. These experiments enable comprehensive evaluations, comparing the performance of Afrocentric models against mainstream PLMs. Our results show that the Afro-XLMR model outperforms all other models, achieving an average F1-score of 71.04% across five tested languages, and the lowest error rate among the evaluated models. Additionally, we enhance the interpretability and explainability of the Afro-XLMR model using Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP). These XAI techniques ensure that sentiment predictions are not only accurate and interpretable but also understandable, fostering trust and reliability in AI-driven NLP technologies, particularly in the context of African languages.

A CNN-accelerated workflow for stochastic seismic property estimation

As one of the major tools in resolving the non-uniqueness challenge in subsurface interpretation and reservoir characterization, stochastic inversion from post- and pre-stack seismic data remains challenging, which not only requires heavy computational resources but also relies on intensive manual supervision. Inspired by the recent advances in deep learning particularly convolutional neural networks (CNNs) for interdisciplinary data integration, this study proposes a deep learning workflow that enables stochastic property estimation by efficiently integrating seismic images with sparse wells. It starts with sampling a set of property prior models (PPMs) from densely-measured properties at well locations and corrupting local seismic patterns with Gaussian noise. The core idea is to train a structure-guided CNN by mapping the contaminated seismic with the sampled PPMs while enforcing structural consistency to avoid overfitting in the presence of sparse wells. Finally, the baseline and uncertainty of target properties are estimated by running multiple realizations of the trained CNN. As demonstrated by three examples, with minimum efforts of CNN architecture customization according to data availability, the proposed workflow can accommodate various use cases, including rock acoustic/elastic property estimation from 3D post-/angle-stack seismic and soil geotechnical properties from 2D ultra-high-resolution seismic. In all examples, the machine predictions match seismic patterns well and are of high lateral consistency.

Optimization of Edge–Cloud Collaborative Computing Resource Management for Internet of Vehicles Based on Multiagent Deep Reinforcement Learning

Optimization of Edge–Cloud Collaborative Computing Resource Management for Internet of Vehicles Based on Multiagent Deep Reinforcement Learning

Lightweight remote sensing super-resolution with multi-scale graph attention network

Remote Sensing Super-Resolution (RS-SR) constitutes a pivotal component in the domain of remote sensing image analysis, aimed at enhancing the spatial resolution of low-resolution imagery. Recent advancements have seen deep learning techniques achieving substantial progress in the RS-SR field. Notably, Graph Neural Networks (GNNs) have emerged as a potent mechanism for processing remote sensing images, adept at elucidating the intricate inter-pixel relationships within images. Nevertheless, a prevalent limitation among existing GNN-based methodologies is their disregard for the high computational demands, which circumscribes their applicability in environments with limited computational resources. This paper introduces a streamlined RS-SR framework, leveraging a Multi-Scale Graph Attention Network (MSGAN), designed to effectively balance computational efficiency with high performance. The core of MSGAN is a novel multi-scale graph attention module, integrating graph attention block and multi-scale lattice block structures, engineered to comprehensively assimilate both localized and extensive spatial information in remote sensing images. This enhances the framework’s overall efficacy and resilience in RS-SR tasks. Comparative experimental analyses demonstrate that MSGAN delivers competitive results against state-of-the-art methods while reducing parameter count and computational overhead, presenting a promising avenue for deployment in scenarios with limited computational resources.

Earthwork Network Architecture (ENA): Research for Earthwork Quantity Estimation Method Improvement with Large Language Model

Accurate earthwork quantity estimation is essential for effective project planning and cost management in the Architecture, Engineering, and Construction (AEC) industry. Traditional methods for quantity takeoff are often time-consuming and susceptible to human error, particularly when working with unstructured datasets such as CAD drawings. This study introduces the Earthwork Network Architecture (ENA), a novel deep learning framework that incorporates Large Language Models (LLMs), Multi-Layer Perceptron (MLP), Long Short-Term Memory (LSTM) networks, and Transformers to automate and enhance the accuracy of earthwork quantity estimation. We assume that if LLMs can be trained effectively using such unstructured construction dataset, the effects such as improved accuracy and the challenges of LLMs can be clearly examined. Among the architectures tested, the LLM-based ENA demonstrated superior performance, achieving faster convergence, greater loss reduction, and higher classification accuracy, with a Quantity Takeoff Classification accuracy of 97.17%. However, the LLMs required significantly more computational resources compared with other models. These findings suggest that LLMs, typically used in natural language processing, can be effectively adapted for complex AEC datasets. This study lays the groundwork for future AI-driven solutions in the AEC industry, underscoring the potential of LLMs and Transformers to automate the quantity takeoff process and manage multimodal data in construction projects.