Random Splits Research Articles

A Novel Machine Learning-Based Predictive Model of Clinically Significant Prostate Cancer and Online Risk Calculator

ObjectivesTo create a machine learning predictive model combining PI-RADS score, PSA density, and clinical variables to predict clinically significant prostate cancer (csPCa). MethodsWe evaluated a cohort of patients who underwent prostate biopsy for suspected prostate cancer (PCa) in New Zealand, Australia, and Switzerland. We collected data on age, body mass index (BMI), PSA level, prostate volume, PSA density (PSAD), PI-RADS scores, previous biopsy, and corresponding histology results. The dataset was divided into derivation (training) and validation (test) sets using random splits. An independent dataset was obtained from the Harvard Dataverse for external validation. A cohort of 1272 patients was analyzed. We fitted a Lasso model, XGBoost, and LightGBM to the training set and assessed their accuracy. ResultsAll models demonstrated ROC AUC values ranging from 0.830 to 0.851. LightGBM was considered the superior model, with an ROC of 0.851 [95%CI: 0.804 – 0.897] in the test set and 0.818 [95% CI: 0.798 – 0.831] in the external dataset. The most important variable was PI-RADS, followed by PSA density, history of previous biopsy, age, and BMI. ConclusionsWe developed a predictive model for detecting csPCa that exhibited a high ROC-AUC value for internal and external validations. This suggests that the integration of the clinical parameters outperformed each individual predictor. Additionally, the model demonstrated good calibration metrics, indicative of a more balanced model than the existing models.

MD-HIT: Machine learning for material property prediction with dataset redundancy control

Materials datasets usually contain many redundant (highly similar) materials due to the tinkering approach historically used in material design. This redundancy skews the performance evaluation of machine learning (ML) models when using random splitting, leading to overestimated predictive performance and poor performance on out-of-distribution samples. This issue is well-known in bioinformatics for protein function prediction, where tools like CD-HIT are used to reduce redundancy by ensuring sequence similarity among samples greater than a given threshold. In this paper, we survey the overestimated ML performance in materials science for material property prediction and propose MD-HIT, a redundancy reduction algorithm for material datasets. Applying MD-HIT to composition- and structure-based formation energy and band gap prediction problems, we demonstrate that with redundancy control, the prediction performances of the ML models on test sets tend to have relatively lower performance compared to the model with high redundancy, but better reflect models’ true prediction capability.

Development of a CVD mortality risk score using nutritional predictors: A risk prediction model in the Golestan Cohort Study

Background and AimWe aimed to develop a dietary score using prediction model method for evaluating the risk of cardiovascular disease (CVD) mortality and suggesting a simple and practical scoring system within the healthcare context. Method and ResultsA total of 43878 adult participants (aged 37-80 years) from the Golestan Cohort Study (GCS) were included in analysis. A random split of the subjects into the derivation (n=28930) and the validation sets (n=14948) was done. The Cox proportional hazard model was used to develop prediction model for the 8-year risk of CVD mortality. The model’s discrimination and calibration were assessed by C-statistic and calibration plot, respectively. To enhance clinical utility, we devised a point-based scoring system derived from our model. This prediction model was developed by nine predictors including age, physical activity level (MET minutes/week), waist-to-hip ratio, tea intake (cup/day), vegetable intake (gr/1000 kcal/day), white meat intake (gr/1000 kcal/day), salt intake (gr/1000 kcal/day), dairy intake (Cup/1000 kcal/day), and percentage of protein intake. The model had an acceptable discrimination in both derivation (C-statistic: 0.76, p<0.001) and validation (C- statistic: 0.77, p<0.001) samples. Also, the calibration of model in both derivation and validation datasets was 0.81. ConclusionThis is the first attempt to develop a risk prediction model of CVD mortality and the risk scoring system by the majority of nutritional predictors in a large cohort study. This nutritional risk assessment tool is suitable for motivating at-risk individuals to make lifestyle and dietary pattern changes to reduce future risk to prevent health problems.

Chemometrics and analytical blank on the at-line monitoring of Zika-VLP production using near-infrared spectroscopy

The Zika disease caused by the Zika virus was declared a Public Health Emergency by the World Health Union (WHO), with microcephaly as the most critical consequence. Aiming to reduce the spread of the virus, biopharmaceutical organizations invest in vaccine research and production, based on multiple platforms. A crescent vaccine production approach is based on virus-like particles (VLP), for not having genetic material in its composition, hypoallergenic and non-mutant character. For bioprocess, it is essential to have means of real-time monitoring, which can be assessed using process analysis techniques such as Near-infrared (NIR) spectroscopy, that can be combined with chemometric methods, like Partial-Least Squares (PLS) and Artificial Neural Networks (ANN) for prediction of biochemical variables. This work proposes a biochemical Zika VLP upstream production at-line monitoring model using NIR spectroscopy comparing sampling conditions (with or without cells), analytical blank (air, ultrapure water), and spectra pre-processing approaches. Seven experiments in a benchtop bioreactor using recombinant baculovirus/Sf9 insect cell platform in serum-free medium were performed to obtain biochemical and spectral data for chemometrics modeling (PLS and ANN), composed by a random data split (80 % calibration, 20 % validation) for cross-validation of the PLS models and 70 % training, 15 % testing, 15 % validation for ANN. The best models generated in the present work presented an average absolute error of 1.59 × 105 cell/mL for density of viable cells, 2.37 % for cell viability, 0.25 g/L for glucose, 0.007 g/L for lactate, 0.138 g/L for glutamine, 0.18 g/L for glutamate, 0,003 g/L for ammonium, and 0.014 g/L for potassium.

Enhanced Osteoporosis Detection Using Artificial Intelligence: A Deep Learning Approach to Panoramic Radiographs with an Emphasis on the Mental Foramen.

Osteoporosis, a skeletal disorder, is expected to affect 60% of women aged over 50 years. Dual-energy X-ray absorptiometry (DXA) scans, the current gold standard, are typically used post-fracture, highlighting the need for early detection tools. Panoramic radiographs (PRs), common in annual dental evaluations, have been explored for osteoporosis detection using deep learning, but methodological flaws have cast doubt on otherwise optimistic results. This study aims to develop a robust artificial intelligence (AI) application for accurate osteoporosis identification in PRs, contributing to early and reliable diagnostics. A total of 250 PRs from three groups (A: osteoporosis group, B: non-osteoporosis group matching A in age and gender, C: non-osteoporosis group differing from A in age and gender) were cropped to the mental foramen region. A pretrained convolutional neural network (CNN) classifier was used for training, testing, and validation with a random split of the dataset into subsets (A vs. B, A vs. C). Detection accuracy and area under the curve (AUC) were calculated. The method achieved an F1 score of 0.74 and an AUC of 0.8401 (A vs. B). For young patients (A vs. C), it performed with 98% accuracy and an AUC of 0.9812. This study presents a proof-of-concept algorithm, demonstrating the potential of deep learning to identify osteoporosis in dental radiographs. It also highlights the importance of methodological rigor, as not all optimistic results are credible.

Machine-Learning Predictions of Critical Temperatures from Chemical Compositions of Superconductors.

In the quest for advanced superconducting materials, the accurate prediction of critical temperatures (Tc) poses a formidable challenge, largely due to the complex interdependencies between superconducting properties and the chemical and structural characteristics of a given material. To address this challenges, we have developed a machine-learning framework that aims to elucidate these complicated and hitherto poorly understood structure-property and property-property relationships. This study introduces a novel machine-learning-based workflow, termed the Gradient Boosted Feature Selection (GBFS), which has been tailored to predict Tc for superconductors by employing a distributed gradient-boosting framework. This approach integrates exploratory data analyses, statistical evaluations, and multicollinearity reduction techniques to select highly relevant features from a high-dimensional feature space, derived solely from the chemical composition of materials. Our methodology was rigorously tested on a data set comprising approximately 16,400 chemical compounds with around 12,000 unique chemical compositions. The GBFS workflow enabled the development of a classification model that distinguishes compositions likely to exhibit Tc values greater than 10 K. This model achieved a weighted average F1-score of 0.912, an AUC-ROC of 0.986, and an average precision score of 0.919. Additionally, the GBFS workflow underpinned a regression model that predicted Tc values with an R2 of 0.945, an MAE of 3.54 K, and an RMSE of 6.57 K on a test set obtained via random splitting. Further exploration was conducted through out-of-sample Tc predictions, particularly those exceeding the liquid nitrogen temperature, and out-of-distribution predictions for (Ca1-xLax)FeAs2 based on varying lanthanum content. The outcome of our study underscores the significance of systematic feature analysis and selection in enhancing predictive model performance, offering various advantages over models that rely primarily on algorithmic complexity. This research not only advances the field of superconductivity but also sets a precedent for the application of machine learning in materials science.

Intelligent Surface Recognition for Autonomous Tractors Using Ensemble Learning with BNO055 IMU Sensor Data

This study aims to enhance the navigation capabilities of autonomous tractors by predicting the surface type they are traversing using data collected from BNO055 Inertial Measurement Units (IMU sensors). IMU sensor data were collected from a small mobile robot driven over seven different floor surfaces within a university environment, including tile, carpet, grass, gravel, asphalt, concrete, and sand. Several machine learning models, including Logistic Regression, K-Neighbors, SVC, Decision Tree, Random Forest, Gradient Boosting, AdaBoost, and XGBoost, were trained and evaluated to predict the surface type based on the sensor data. The results indicate that Random Forest and XGBoost achieved the highest accuracy, with scores of 98.5% and 98.7% in K-Fold Cross-Validation, respectively, and 98.8% and 98.6% in an 80/20 Random State split. These findings demonstrate that ensemble methods are highly effective for this classification task. Accurately identifying surface types can prevent operational errors and improve the overall efficiency of autonomous systems. Integrating these models into autonomous tractor systems can significantly enhance adaptability and reliability across various terrains, ensuring safer and more efficient operations.

Machine learning approaches identify immunologic signatures of total and intact HIV DNA during long-term antiretroviral therapy.

Understanding the interplay between the HIV reservoir and the host immune system may yield insights into HIV persistence during antiretroviral therapy (ART) and inform strategies for a cure. Here, we applied machine learning (ML) approaches to cross-sectional high-parameter HIV reservoir and immunology data in order to characterize host-reservoir associations and generate new hypotheses about HIV reservoir biology. High-dimensional immunophenotyping, quantification of HIV-specific T cell responses, and measurement of genetically intact and total HIV proviral DNA frequencies were performed on peripheral blood samples from 115 people with HIV (PWH) on long-term ART. Analysis demonstrated that both intact and total proviral DNA frequencies were positively correlated with T cell activation and exhaustion. Years of ART and select bifunctional HIV-specific CD4 T cell responses were negatively correlated with the percentage of intact proviruses. A leave-one-covariate-out inference approach identified specific HIV reservoir and clinical-demographic parameters, such as age and biological sex, that were particularly important in predicting immunophenotypes. Overall, immune parameters were more strongly associated with total HIV proviral frequencies than intact proviral frequencies. Uniquely, however, expression of the IL-7 receptor alpha chain (CD127) on CD4 T cells was more strongly correlated with the intact reservoir. Unsupervised dimension reduction analysis identified two main clusters of PWH with distinct immune and reservoir characteristics. Using reservoir correlates identified in these initial analyses, decision tree methods were employed to visualize relationships among multiple immune and clinical-demographic parameters and the HIV reservoir. Finally, using random splits of our data as training-test sets, ML algorithms predicted with approximately 70% accuracy whether a given participant had qualitatively high or low levels of total or intact HIV DNA . The techniques described here may be useful for assessing global patterns within the increasingly high-dimensional data used in HIV reservoir and other studies of complex biology.

Machine learning approaches identify immunologic signatures of total and intact HIV DNA during long-term antiretroviral therapy

Understanding the interplay between the HIV reservoir and the host immune system may yield insights into HIV persistence during antiretroviral therapy (ART) and inform strategies for a cure. Here, we applied machine learning (ML) approaches to cross-sectional high-parameter HIV reservoir and immunology data in order to characterize host–reservoir associations and generate new hypotheses about HIV reservoir biology. High-dimensional immunophenotyping, quantification of HIV-specific T cell responses, and measurement of genetically intact and total HIV proviral DNA frequencies were performed on peripheral blood samples from 115 people with HIV (PWH) on long-term ART. Analysis demonstrated that both intact and total proviral DNA frequencies were positively correlated with T cell activation and exhaustion. Years of ART and select bifunctional HIV-specific CD4 T cell responses were negatively correlated with the percentage of intact proviruses. A leave-one-covariate-out inference approach identified specific HIV reservoir and clinical–demographic parameters, such as age and biological sex, that were particularly important in predicting immunophenotypes. Overall, immune parameters were more strongly associated with total HIV proviral frequencies than intact proviral frequencies. Uniquely, however, expression of the IL-7 receptor alpha chain (CD127) on CD4 T cells was more strongly correlated with the intact reservoir. Unsupervised dimension reduction analysis identified two main clusters of PWH with distinct immune and reservoir characteristics. Using reservoir correlates identified in these initial analyses, decision tree methods were employed to visualize relationships among multiple immune and clinical–demographic parameters and the HIV reservoir. Finally, using random splits of our data as training-test sets, ML algorithms predicted with approximately 70% accuracy whether a given participant had qualitatively high or low levels of total or intact HIV DNA . The techniques described here may be useful for assessing global patterns within the increasingly high-dimensional data used in HIV reservoir and other studies of complex biology.

Explainable Ensemble Learning and Multilayer Perceptron Modeling for Compressive Strength Prediction of Ultra-High-Performance Concrete.

The performance of ultra-high-performance concrete (UHPC) allows for the design and creation of thinner elements with superior overall durability. The compressive strength of UHPC is a value that can be reached after a certain period of time through a series of tests and cures. However, this value can be estimated by machine-learning methods. In this study, multilayer perceptron (MLP) and Stacking Regressor, an ensemble machine-learning models, is used to predict the compressive strength of high-performance concrete. Then, the ML model's performance is explained with a feature importance analysis and Shapley additive explanations (SHAPs), and the developed models are interpreted. The effect of using different random splits for the training and test sets has been investigated. It was observed that the stacking regressor, which combined the outputs of Extreme Gradient Boosting (XGBoost), Category Boosting (CatBoost), Light Gradient Boosting Machine (LightGBM), and Extra Trees regressors using random forest as the final estimator, performed significantly better than the MLP regressor. It was shown that the compressive strength was predicted by the stacking regressor with an average R2 score of 0.971 on the test set. On the other hand, the average R2 score of the MLP model was 0.909. The results of the SHAP analysis showed that the age of concrete and the amounts of silica fume, fiber, superplasticizer, cement, aggregate, and water have the greatest impact on the model predictions.

Predicting the performance of lithium adsorption and recovery from unconventional water sources with machine learning

Selective lithium (Li) recovery from unconventional water sources (UWS) (e.g., shale gas waters, geothermal brines, and rejected seawater desalination brines) using inorganic lithium-ion sieve (LIS) materials can address Li supply shortages and distribution issues. However, the development of high-performance LIS materials and the optimization of recovery-related operating parameters are hampered by the variety of production methods, intricate procedures, and experimental expenses. Machine learning (ML) techniques offer potential solutions for enhancing LIS material development. We collected literature data on Li adsorption, categorizing 16 parameters into adsorbent parameters, operating parameters, and solution components. Three tree-based algorithms—Random Forest (RF), Gradient Boosting Decision Trees (GBDT), and Extreme Gradient Boosting (XGBoost)—were used to evaluate the impact of these parameters on lithium adsorption. The grouped random splitting method limited data leakage and mitigated overfitting. XGBoost demonstrated the best performance, with an R² of 0.98 and a root-mean-squared error (RMSE) of 1.72. The SHAP values highlighted that operating parameters were the most influential, followed by adsorbent parameters and coexisting ion concentrations. Therefore, focusing on optimizing operating parameters or making targeted improvements on LIS based on operating conditions will enhance LIS performances in UWS. These insights are crucial for optimizing Li adsorption processes and designing effective inorganic LIS materials.

Artificial neural network prediction of postoperative complications in papillary thyroid microcarcinoma based on preoperative ultrasonographic features.

To predict post-thyroidectomy complications in papillary thyroid microcarcinoma (PTMC) patients using a deep learning model based on preoperative ultrasonographic features. This study addresses the global rise in PTMC incidence and the challenges in treatment decision-making with high-resolution ultrasonography. This study enrolled 1638 patients with clinically staged cN0 PTMC who received surgical treatment from 1997 to 2019 at Beijing Friendship Hospital. Deep learning model was developed using fully connected neural network. Feature selection included 1000 iterations of Bootstrap sampling and Recursive Feature Elimination (RFE) to identify the top 10 features. Data preprocessing involved normalization and imputation for missing values. SMOTE addressed class imbalance. The model was trained and tested on random data split, with performance metrics including Accuracy (ACC), Area Under the Curve (AUC), Sensitivity (SEN), and Specificity (SPE), visualized through a ROC curve and confusion matrix. The fully connected deep neural network model demonstrated high accuracy (ACC 0.81), Area Under the Curve (AUC 0.74), sensitivity (SEN 0.65), and specificity (SPE 0.83) and visualized by ROC curve and confusion matrix. These results highlight the model's reliability and potential as an effective tool in predicting postoperative complications and assisting in clinical decision-making for PTMC patients. This study highlights the potential of deep learning in enhancing medical predictions and personalized healthcare. Despite promising results, limitations include a single-center data source and unconsidered factors like lifestyle and genetics. Future research should expand data sources, include more influencing factors, and refine algorithms to improve accuracy and applicability in thyroid cancer treatment.

Using a random cross-validation technique to compare typical regression vs. Random Forests for modelling pan evaporation

Pan evaporation (Epan) of class A pan evaporimeter under local semi-arid conditions was modelled in this study based on meteorological observations as input data using an integrated regression approach that includes three steps: a) first step: appropriate selection of transformations for reducing normality departures of independent variables and ridge regression for selecting variables with low collinearity based on variance inflation factors, b) second step (RCV-REG): regression (REG) of the final model with selected transformed variables of low collinearity implemented using an iterative procedure called “Random Cross-Validation” (RCV) that splits multiple times the data in calibration and validation subsets considering a random selection procedure, c) robustness control of the estimated regression coefficients from RCV-REG by analyzing the sign (+ or -) variation of their iterative solutions using the 95% interval of their Highest Posterior Density Distribution (HPD). The iterative procedure of RCV can also be implemented on machine learning methods (MLs) and for this reason, the ML method of Random Forests (RF) was also applied with RCV (RCV-RF) as an additional case in order to be compared with RCV-REG. Random splitting of data into calibration and validation set (70% and 30%, respectively) was performed 1,000 times in RCV-REG and led to a respective number of solutions of the regression coefficients. The same number of iterations and random splitting for validation was also used in the RCV-RF. The results showed that RCV-REG outperformed RCV-RF at all model performance criteria providing robust regression coefficients associated to independent variables (constant signs of their 95% HPD interval) and better distribution of validation solutions in the iterative 1:1 plots from RCV-RF (RCV-RG: R2=0.843, RMSE=0.853, MAE=0.642, MAPE=0.081, NSE=0.836, Slope(1:1 plot)=0.998, Intercept(1:1 plot)=0.011, and RCV-RF: R2=0.835, RMSE=0.904, MAE=0.689, MAPE= 0.088, NSE=0.818, Slope(1:1 plot)=1.120, Intercept(1:1 plot)=-1.011, based on the mean values of 1,000 iterations). The use of RCV approach in various modelling approaches solves the problem of subjective splitting of data into calibration and validation sets, provides a better evaluation of the final modelling approaches and enhances the competitiveness of typical regression models against machine learning models.

Multicellular target QSAR models for predicting of novel inhibitors against pancreatic cancer by Monte Carlo approach

Pancreatic cancer is one of the most fatal malignant tumors, with a poor prognosis. Nowadays, computational methods for designing and predicting anticancer agents have gained increased attention. In this work, a QSAR study is conducted to further investigate the activity of 5797 agents against pancreatic cancer cell lines. QSAR models are a system of classification approaches that are used to construct and validate the predictive potential of created models. This approach demonstrates the prediction in the active (1) and inactive (−1) format. Here, the classification QSAR technique is employed to construct and evaluate models. The classification QSAR model has been constructed by the Monte Carlo method based on the representation of the molecular structure by SMILES (simplified molecular input line entry systems) using the CORAL software. Four random splits of data by making four sets such as active training, passive training, calibration, and validation sets are assessed. The numerical values of sensitivity, specificity, and accuracy for the validation set of splits 1 to 4 are in the range of 0.7726–0.7901, 0.9007–0.9220, and 0.8548–0.8730 respectively. The Matthews correlation coefficient (MCC) of these models for external validation sets is found to be in the range of 0.6832 to 0.7215.

Machine learning potentials for global multi-timescale diffuse irradiance estimation: Synthesizing ground observations, time-series, and environmental features

Separating diffuse horizontal irradiance (DHI) from ground-based global horizontal irradiance observations is critical owing to lack of direct DHI observations. Existing models are often site-specific and time-bound, thereby limiting their universal applicability. To address this, this study thoroughly explores machine learning (ML) for constructing global separation models. We develop 36 models using three ML algorithms—light gradient boosting machine (LightGBM), generalized additive model (GAM), and geographically weighted artificial neural network (GWANN)—alongside three data splitting strategies for minutely, 10-minutely, hourly, and daily timescales. LightGBM and GAM outperform GWANN in accuracy, with LightGBM excelling in interpretability, efficiency, and handling missing values, while GWANN exhibits superior stability. Compared with completely random splitting, model accuracy decreases by 2.26 % and 2.16 % for station- and date-based splitting, respectively. Prediction accuracy varies across timescales, with minutely models typically considered the best. Key predictors are Kt, solar zenith angle, and altitude, complemented by the significant influence of time-series Kt, aerosol, and cloud features. The influence of various factors on model accuracy differs and is scale-dependent. Among them, time-series Kt variability factors notably enhance ML-based predictions, especially at minutely scales. The LightGBM model outperforms classic models in overall and site-specific accuracy and adaptability, but its computational efficiency declines, especially with time-series variability factors. The findings highlight that ML-based separation models necessitate careful selection of algorithm, data splitting strategy, and input variables, fully considering study region, data situation, and temporal and spatial scales. This research offers a potential solution for constructing global separation models and insights for model improvement.

Estimating Brazilian Amazon Canopy Height Using Landsat Reflectance Products in a Random Forest Model with Lidar as Reference Data

Landsat data have been used to derive forest canopy structure, height, and volume using machine learning models, i.e., giving computers the ability to learn from data and make decisions and predictions without being explicitly programmed, with training data provided by ground measurement or airborne lidar. This study explored the potential use of Landsat reflectance and airborne lidar data as training data to estimate canopy heights in the Brazilian Amazon forest and examined the impacts of Landsat reflectance products at different process levels and sample spatial autocorrelation on random forest modeling. Specifically, this study assessed the accuracy of canopy height predictions from random forest regression models impacted by three different Landsat 8 reflectance product inputs (i.e., USGS level 1 top of atmosphere reflectance, USGS level 2 surface reflectance, and NASA nadir bidirectional reflectance distribution function (BRDF) adjusted reflectance (NBAR)), sample sizes, training/test split strategies, and geographic coordinates. In the establishment of random forest regression models, the dependent variable (i.e., the response variable) was the dominant canopy heights at a 90 m resolution derived from airborne lidar data, while the independent variables (i.e., the predictor variables) were the temporal metrics extracted from each Landsat reflectance product. The results indicated that the choice of Landsat reflectance products had an impact on model accuracy, with NBAR data yielding more trustful results than the other products despite having higher RMSE values. Training and test split strategy also affected the derived model accuracy metrics, with the random sample split (randomly distributed training and test samples) showing inflated accuracy compared to the spatial split (training and test samples spatially set apart). Such inflation was induced by the spatial autocorrelation that existed between training and test data in the random split. The inclusion of geographic coordinates as independent variables improved model accuracy in the random split strategy but not in the spatial split, where training and test samples had different geographic coordinate ranges. The study highlighted the importance of data processing levels and the training and test split methods in random forest modeling of canopy height.

Spontaneous symmetry breaking of coupled Fabry–Pérot nanocavities

Ring cavities are quickly emerging as preferential platforms to investigate spontaneous symmetry breaking. However, research on handedness polarisation splitting within Kerr-coupled cavities is yet at a preliminary stage. Here, we observe experimentally spontaneous symmetry breaking in a non-Hermitian coupled Fabry‒Pérot nanocavity. In the experiment, horizontally polarised light is incident on the nanocavities, and symmetry breaking occurs when the power exceeds the symmetry breaking threshold of 8mW\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$8\\,{{\\mbox{mW}}}$$\\end{document}. We interpret such observation via nonlinear coupled mode theory, finding that an excitation power-dependent random splitting of right- or left-handed circular polarisations (RCP and LCP, respectively) emerges at the resonance peak. Further numerical simulations show that when the incident power is above the symmetry breaking threshold, the device exhibits spontaneous symmetry breaking characteristics: an additional polarisation component appears in the output field when the input field is monopolarised, and this is attributed to the imbalance of RCP and LCP. Our findings provide further understanding of spontaneous symmetry breaking in non-Hermitian systems and demonstrate the potential applications of the proposed device in optical signal processing.

Statistical Inference for Heterogeneous Treatment Effects Discovered by Generic Machine Learning in Randomized Experiments

Researchers are increasingly turning to machine learning (ML) algorithms to investigate causal heterogeneity in randomized experiments. Despite their promise, ML algorithms may fail to accurately ascertain heterogeneous treatment effects under practical settings with many covariates and small sample size. In addition, the quantification of estimation uncertainty remains a challenge. We develop a general approach to statistical inference for heterogeneous treatment effects discovered by a generic ML algorithm. We apply the Neyman’s repeated sampling framework to a common setting, in which researchers use an ML algorithm to estimate the conditional average treatment effect and then divide the sample into several groups based on the magnitude of the estimated effects. We show how to estimate the average treatment effect within each of these groups, and construct a valid confidence interval. In addition, we develop nonparametric tests of treatment effect homogeneity across groups, and rank-consistency of within-group average treatment effects. The validity of our methodology does not rely on the properties of ML algorithms because it is solely based on the randomization of treatment assignment and random sampling of units. Finally, we generalize our methodology to the cross-fitting procedure by accounting for the additional uncertainty induced by the random splitting of data.

Structure-based out-of-distribution (OOD) materials property prediction: a benchmark study

In real-world materials research, machine learning (ML) models are usually expected to predict and discover novel exceptional materials that deviate from the known materials. It is thus a pressing question to provide an objective evaluation of ML model performances in property prediction of out-of-distribution (OOD) materials that are different from the training set. Traditional performance evaluation of materials property prediction models through the random splitting of the dataset frequently results in artificially high-performance assessments due to the inherent redundancy of typical material datasets. Here we present a comprehensive benchmark study of structure-based graph neural networks (GNNs) for extrapolative OOD materials property prediction. We formulate five different categories of OOD ML problems for three benchmark datasets from the MatBench study. Our extensive experiments show that current state-of-the-art GNN algorithms significantly underperform for the OOD property prediction tasks on average compared to their baselines in the MatBench study, demonstrating a crucial generalization gap in realistic material prediction tasks. We further examine the latent physical spaces of these GNN models and identify the sources of CGCNN, ALIGNN, and DeeperGATGNN’s significantly more robust OOD performance than those of the current best models in the MatBench study (coGN and coNGN) as a case study for the perovskites dataset, and provide insights to improve their performance.

Evaluation of Immediate and Delayed Placement of Post on Sealing ability of Periapical Area: An Original Research

ABSTRACT Objective: The purpose of this research is to compare the effects of post placement—immediate and delayed—on the periapical area’s capacity to close after root canal therapy (RCT). Techniques: A random split of sixty recently removed human single-rooted teeth was made into two groups: Group A received immediate post placement, whereas Group B received post placement after seven days. Standard guidelines were followed in the preparation of root canals and the placement of posts. A dye penetration method was used to measure microleakage, a measure of sealing capacity. Findings: Compared to the immediate post placement group (0.62 ± 0.08) (P &lt; 0.001), the delayed post placement group had a considerably lower mean microleakage score (0.35 ± 0.06), showing improved sealing capacity. In conclusion, post implantation done later enhances the periapical area’s sealing capacity after RCT. This research underscores the significance of scheduling endodontic treatments to maximize treatment success and improve tooth retention, as well as the possible therapeutic advantages of postponing post implantation.