In situ fatigue crack propagation experiment was conducted on laser cladding with coaxial powder feeding (LCPF) K477 under various stress ratios and temperatures. Multiple crack initiation sites were observed by using...
详细信息
In situ fatigue crack propagation experiment was conducted on laser cladding with coaxial powder feeding (LCPF) K477 under various stress ratios and temperatures. Multiple crack initiation sites were observed by using in situ scanning electron microscopy (SEM). The fatigue short crack growth rate was measured, and the impacts of temperature and stress ratio on this growth rate were analyzed. Based on these experiments, the experimental data were expanded, and three ensemble learning algorithms, that is, random forest (RF), extreme gradient boosting (XGBoost), and light gradient boosting machine (LightGBM), were employed to establish a fatigue short crack growth rate model controlled by multiple parameters. It is indicated that the RF model performs the best, achieving a coefficient of determination (R2) of up to 0.88. The fatigue life predicted by the machinelearning (ML) method agrees well with the experimental one.
The gastrointestinal (GI) tract plays a crucial role in the human body but is frequently affected by diseases that pose a significant global health issue. GI endoscopy is the primary diagnostic tool for early disease ...
详细信息
The local structure and thermophysical behavior of Mg-La liquid alloys were in-depth understood using deep potential molecular dynamic(DPMD) simulation driven via machinelearning to promote the development of Mg-La a...
详细信息
The local structure and thermophysical behavior of Mg-La liquid alloys were in-depth understood using deep potential molecular dynamic(DPMD) simulation driven via machinelearning to promote the development of Mg-La alloys. The robustness of the trained deep potential(DP) model was thoroughly evaluated through several aspects, including root-mean-square errors(RMSEs), energy and force data, and structural information comparison results;the results indicate the carefully trained DP model is reliable. The component and temperature dependence of the local structure in the Mg-La liquid alloy was analyzed. The effect of Mg content in the system on the first coordination shell of the atomic pairs is the same as that of temperature. The pre-peak demonstrated in the structure factor indicates the presence of a medium-range ordered structure in the Mg-La liquid alloy, which is particularly pronounced in the 80at% Mg system and disappears at elevated temperatures. The density, self-diffusion coefficient, and shear viscosity for the Mg-La liquid alloy were predicted via DPMD simulation, the evolution patterns with Mg content and temperature were subsequently discussed, and a database was established accordingly. Finally, the mixing enthalpy and elemental activity of the Mg-La liquid alloy at 1200 K were reliably evaluated,which provides new guidance for related studies.
With the rapid growth of the courier industry driven by societal development, efficiently and accurately predicting the send-receive path has become a critical issue. Existing methods suffer from several limitations, ...
详细信息
With the rapid growth of the courier industry driven by societal development, efficiently and accurately predicting the send-receive path has become a critical issue. Existing methods suffer from several limitations, including discrepancies between pretraining and downstream data, and inconsistencies between training and testing targets. To address these challenges, this paper introduces a pioneering approach that applies prompt learning to send-receive path prediction. The proposed method employs a "pretraining-prompt-finetuning" paradigm, where a model is pretrained on a large-scale dataset and then finetuned using prompt vectors to adapt to downstream tasks. This novel strategy effectively bridges the gap between pretraining and finetuning data, ensuring better model generalization with minimal additional cost. Furthermore, we incorporate an actor-critic reinforcement learning framework, where the actor network generates paths and the critic network evaluates them. This framework optimizes the model based on rewards calculated from non-differentiable test criteria, effectively addressing the inconsistency between training and testing objectives. This approach is better suited to adapt to various delivery scenarios, enhancing prediction accuracy and efficiency. Experiments conducted on two real-world datasets demonstrate the superiority of the proposed method.
Accurately and rapidly simulating the hysteretic behavior of double skin composite wall (DSCW) under earthquake loads enhance the efficiency of seismic performance assessments for high-rise steel-concrete composite st...
详细信息
Accurately and rapidly simulating the hysteretic behavior of double skin composite wall (DSCW) under earthquake loads enhance the efficiency of seismic performance assessments for high-rise steel-concrete composite structures. The application of artificial intelligence in civil engineering enables a shift from purely physics-based models to methods that integrate physical mechanisms with data-driven approaches. To this end, a machinelearning-aided model for simulating the hysteretic behavior of DSCW under earthquake loads is proposed. Firstly, a fiber-based uniaxial material model incorporating softening was proposed to replicate the primary damaging mechanisms of DSCW, validated through comparison with experimental data. A comprehensive dataset covering a broad range of engineering metrics of DSCW was generated, incorporating various loading protocols. The modified Bouc-Wen model was utilized to calibrate hysteretic curves, solving an inverse problem via optimization techniques to determine consistent hysteretic parameters (CHPs) of DSCW. The analysis results confirm the highest accuracy of the artificial neural network (ANN) model. The SHapley Additive exPlanations (SHAP) analysis provided interpretability, highlighting the significant influence of geometric parameters and confirming the interdependence among Bouc-Wen model parameters. The performance of the ANN model was evaluated against a Transformer-based end-to-end deep learning model. The results indicated that the ANN model exhibited superior performance, particularly under asymmetric loading conditions, due to its integration of physical mechanisms into the force-displacement constitutive relationship.
Knowledge extraction from information stored in databases is always subject to the presence of missing values. Missing data is an unavoidable problem that affects many disciplines of researchers and data scientists. I...
详细信息
ISBN:
(纸本)9783031779404;9783031779411
Knowledge extraction from information stored in databases is always subject to the presence of missing values. Missing data is an unavoidable problem that affects many disciplines of researchers and data scientists. Inasmuch as machinelearning algorithms cannot work with incomplete data in the data sets, data imputation is an essential task to obtain quality data. This research approach provides an overview of the data missingness mechanism and the process of generating synthetic missing data, the imputation of all types of variables, and the performance assessment of several imputation methods. Traditional algorithms, machinelearning methods and various Autoencoder-based deep learning architectures have been studied. An exhaustive analysis and comparison of 21 heterogeneous data sets in various areas has been proposed. They have been exposed to a perturbation procedure with different missingness mechanisms and various missingness rates, covering the different possibilities that can occur in real life. The experimental results show that deep learning models outperform the other methods studied. Furthermore, the performance of data imputation methods does not depend on the missingness mechanism or the synthetic missingness generation method used nor on the percentage of missing values.
The proceedings contain 324 papers. The topics discussed include: sentiment analysis of self-driving car dataset: a comparative study of deep learning approaches;unleashing the potential of boosting techniques to opti...
The proceedings contain 324 papers. The topics discussed include: sentiment analysis of self-driving car dataset: a comparative study of deep learning approaches;unleashing the potential of boosting techniques to optimize station-pairs passenger flow forecasting;a cross-platform movie filtering and recommendation system using big data analytics;a comparative analysis of advanced machinelearning algorithms to diagnose Parkinson’s disease;exploring twitter sentiments for predicting match outcomes in the game of cricket;economic order quantity models with exponential demand rate and single level trade credit;identification of brain diseases using image classification: a deep learning approach;and a short review for handwritten math expression recognition techniques.
Federated learning has emerged as a promising approach to train machinelearning models on decentralized data sources while preserving data privacy. This paper proposes a new federated approach for Naive Bayes (NB) cl...
详细信息
ISBN:
(纸本)9783031777370;9783031777387
Federated learning has emerged as a promising approach to train machinelearning models on decentralized data sources while preserving data privacy. This paper proposes a new federated approach for Naive Bayes (NB) classification, assuming discrete variables. Our approach federates a discriminative variant of NB, sharing meaningless parameters instead of conditional probability tables. Therefore, this process is more reliable against possible attacks. We conduct extensive experiments on 12 datasets to validate the efficacy of our approach, comparing federated and non-federated settings. Additionally, we benchmark our method against the generative variant of NB, which serves as a baseline for comparison. Our experimental results demonstrate the effectiveness of our method in achieving accurate classification.
This paper presents a comparative study of machinelearning models for detecting abusive messages, focusing on code-mixed data in Wolof and French languages. With the increasing use of digital platforms, there has bee...
详细信息
The interaction between soils and geosynthetics plays an important role in the applications of these materials for reinforcement in geotechnical engineering. The complexities of soil-geosynthetic interactions vary dep...
详细信息
The interaction between soils and geosynthetics plays an important role in the applications of these materials for reinforcement in geotechnical engineering. The complexities of soil-geosynthetic interactions vary depending on the type and properties of both the geosynthetic and the soil. This paper introduces a machinelearning approach, specifically a random forest algorithm, for predicting interface friction angles. The dataset comprises 495 interfaces involving geomembranes and sand, with fourteen influencing parameters recorded for each interface, influencing the shear strength outcome. In the analysis, Pearson's correlation coefficient is employed to measure the linear interdependence between each pair of input-input and input-output variables. Following the linear regression analysis, an optimized random forest is utilized to project the interface friction angle. The random forest algorithm divides the selected data into training and testing sets, and only 3% of the training set and 6% of the testing set exceed +/- 5 degrees from the actual records. The coefficient of determination (R-2) indicates strong agreement between the predicted and laboratory study friction angles, with R-2 = 0.93 for the training set and R-2 = 0.92 for the testing set. Consequently, the random forest algorithm demonstrates effectiveness in predicting interface friction angles.
暂无评论