Model Predictions Research Articles

Optimising groundwater recharge to counteract seawater intrusion through an inverse modelling approach and identifying efficient recharge structure locations

In semi-arid coastal urban regions, groundwater faces challenges from seawater intrusion due to reduced recharge and increased extraction. This study employs advanced numerical modeling to comprehensively assess seawater intrusion along the South Chennai coast from 2018 to 2022. The inverse modelling approach utilizes observed data, such as groundwater heads or concentrations, to estimate aquifer properties, like hydraulic conductivity, by minimizing the disparity between model predictions and actual measurements to identify the best-fit parameter values. This inverse method forecasts the necessary recharge rate to address seawater intrusion by increasing groundwater head and decreasing chloride concentrations. The estimation of recharge incorporates variables such as rainfall, soil infiltration capacity, land use, and the vital groundwater draft calculated based on per capita consumption and population. The study uses FEFLOW software to develop a robust conceptual model, simulating groundwater flow and chloride transport. The inverse modeling approach resulted in critical recharge adjustments, enhancing rates by 40%-60% in specific regions. The predictive scenarios, integrating proposed recharge structures, show promising results, notably with a 50% reduction in chloride concentration during the post-monsoon season compared to the pre-monsoon period in wells 8 and 3. A notable decrease in chloride concentration of over 90% was observed during May 2025 in well no: 15 due to its proximity to the Adayar River. These fluctuations are influenced by seasonal variations resulting from climatic changes in South India, with intense rainfall accompanying northeast monsoon winds playing a significant role. Increased recharge rates and strategically placed structures along the Buckingham Canal provide significant relief in the most impacted northern coastal area, acting as effective barriers. These interventions successfully reduce chloride concentration, providing a sustainable and effective solution to mitigate the persistent challenge of seawater intrusion.

A climatic suitability indicator to support Leishmania infantum surveillance in Europe: a modelling study

Leishmaniases are neglected diseases transmitted by sand flies. They disproportionately affect vulnerable groups globally. Understanding the relationship between climate and disease transmission allows the development of relevant decision-support tools for public health policy and surveillance. The aim of this modelling study was to develop an indicator that tracks climatic suitability for Leishmania infantum transmission in Europe at the subnational level. Historical records of sand fly vectors, human leishmaniasis, bioclimatic indicators, and environmental variables were integrated in a machine learning framework (XGBoost) to predict suitability in two past periods (2001-2010 and 2011-2020). We further assessed if predictions were associated with human and animal disease data from selected countries (France, Greece, Italy, Portugal, and Spain). An increase in the number of climatically suitable regions for leishmaniasis was detected, especially in southern and eastern countries, coupled with a northward expansion towards central Europe. The final model had excellent predictive ability (AUC=0.970 [0.947-0.993]), and the suitability predictions were positively associated with human leishmaniasis incidence and canine seroprevalence for Leishmania. This study demonstrates how key epidemiological data can be combined with open-source climatic and environmental information to develop an indicator that effectively tracks spatiotemporal changes in climatic suitability and disease risk. The positive association between the model predictions and human disease incidence demonstrates that this indicator could help target leishmaniasis surveillance to transmission hotspots. European Union Horizon Europe Research and Innovation Programme (European Climate-Health Cluster), United Kingdom Research and Innovation.

Improving Kui digit recognition through machine learning and data augmentation techniques

Speech digit recognition research is growing decisively, and a bulk of digit recognition algorithms are used in European and a few Asian languages. Kui is a low-resourced tribal language locally used in several states of India. Despite its significance, there is not much research on Kui's speech. This research aims to present an in-depth analysis of novel Kui digit recognition using predefined machine learning (ML) techniques. For this purpose, we first gathered spoken numbers i.e. from 0 to 9 of eight different speakers containing a total of 200 words. Secondly, we choose the numbers: ଶୂନ (zero), ଏକ (one), ଦୁଇ (two), ତିନି(three), ସାରି(four), ପାସ (five), ସଅ (six), ସାତ (seven), ଆଟ (eight), ନଅ (nine). Meanwhile, we build nine different ML models to recognize Kui digits that take the Mel-frequency cepstral coefficients (MFCCs) method to extract the relevant features for model predictions. Finally, we compared the performance of ML models for both augmented and non-augmented Kui data. The result shows that the SVM+Augmentation method for Kui digit recognition combined obtained the highest accuracy of 83% than other methods. Moreover, the difficulties and potential prospects for Kui digit recognition are also highlighted in this work.

No dose adjustment of metformin or substrates of organic cation transporters (OCT)1 and OCT2 and multidrug and toxin extrusion protein (MATE)1/2K with fostemsavir coadministration based on modeling approaches.

Fostemsavir is an approved gp120-directed attachment inhibitor and prodrug for the treatment of human immunodeficiency virus type 1 infection in combination with other antiretrovirals (ARVs) in heavily treatment-experienced adults with multi-drug resistance, intolerance, or safety concerns with their current ARV regimen. Initial invitro studies indicated that temsavir, the active moiety of fostemsavir, and its metabolites, inhibited organic cation transporter (OCT)1, OCT2, and multidrug and toxin extrusion transporters (MATEs) at tested concentration of 100 uM, although risk assessment based on the current Food and Drug Administration invitro drug-drug interaction (DDI) guidance using the mechanistic static model did not reveal any clinically relevant inhibition on OCTs and MATEs. However, a DDI risk was flagged with EMA static model predictions. Hence, a physiologically based pharmacokinetic (PBPK) model of fostemsavir/temsavir was developed to further assess the DDI risk potential of OCT and MATEs inhibition by temsavir and predict changes in metformin (a sensitive OCT and MATEs substrate) exposure. No clinically relevant impact on metformin concentrations across a wide range of temsavir concentrations was predicted; therefore, no dose adjustment is recommended for metformin when co-administered with fostemsavir.

Enhancing Breast Cancer Detection: Leveraging Convolutional Neural Networks

Abstract: Breast cancer continues to be a critical health concern, necessitating early detection and accurate classification for effective treatment. This study presents a comparative analysis between a custom-designed Convolutional Neural Network (CNN) and the pre-trained DenseNet121 model for breast cancer detection and classification. We compiled a comprehensive dataset of breast cancer images and applied appropriate preprocessing techniques to optimize the input for the models. The dataset was divided into training, validation, and testing sets to evaluate the models' performance. The CNN model comprises multiple convolutional and pooling layers followed by fully connected layers for classification, while the DenseNet121 model is fine-tuned specifically for breast cancer detection. The models were evaluated based on metrics such as accuracy, precision, recall, F1 score, and AUC-ROC. The DenseNet121 model outperformed the custom CNN model, achieving higher accuracy and reliability. For wider accessibility, we integrated the superior DenseNet121 model into a user-friendly web-based interface using Python Flask, enabling real-time breast cancer predictions. Ethical considerations were paramount, ensuring data privacy, security, and transparency in all model predictions. This study highlights the effectiveness of the DenseNet121 model and contributes to improved breast cancer diagnosis and patient care.

Deep learning in high-frequency trading: Conceptual challenges and solutions for real-time fraud detection

High-frequency trading (HFT) has transformed financial markets by enabling rapid execution of trades, exploiting market inefficiencies, and optimizing trading strategies. However, this speed and complexity also present significant challenges for real-time fraud detection. Deep learning, a subset of machine learning, offers promising solutions to these challenges through its ability to analyze large volumes of data and uncover intricate patterns. This review explores the conceptual challenges and solutions associated with deploying deep learning for fraud detection in HFT environments. One of the primary challenges in implementing deep learning for HFT fraud detection is the sheer volume and velocity of data. HFT systems generate vast amounts of transactional data in milliseconds, necessitating highly efficient and scalable deep learning models. Convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are particularly suited for this task due to their ability to process and analyze sequential data efficiently. However, these models require substantial computational resources and sophisticated infrastructure to operate in real time. Another significant challenge is the need for high accuracy and low latency in fraud detection. False positives can lead to unnecessary interventions, while false negatives can result in undetected fraudulent activities. Deep learning models must be fine-tuned to balance these risks, employing techniques such as hyperparameter optimization and ensemble learning to enhance their predictive capabilities. Additionally, integrating real-time anomaly detection methods can help identify suspicious activities promptly, reducing the window of opportunity for fraudsters. Data quality and integrity also pose substantial challenges. HFT environments are susceptible to noise and outliers, which can distort model predictions. Ensuring high-quality data through rigorous preprocessing and anomaly filtering is crucial for the accuracy of deep learning models. Techniques such as data augmentation and normalization can further improve model robustness. To address these challenges, a hybrid approach combining deep learning with traditional statistical methods and rule-based systems can be effective. This approach leverages the strengths of each method, providing a comprehensive fraud detection framework that is both accurate and responsive. Additionally, ongoing model retraining and adaptation to evolving fraud patterns are essential to maintain the effectiveness of the system. In conclusion, while deep learning presents significant opportunities for enhancing real-time fraud detection in high-frequency trading, it also requires addressing challenges related to data volume, computational demands, accuracy, and data quality. By employing a hybrid approach and continually refining models, financial institutions can effectively mitigate fraud risks and safeguard their trading operations.

Learning From Aggregated Opinion.

The capacity to leverage information from others' opinions is a hallmark of human cognition. Consequently, past research has investigated how we learn from others' testimony. Yet a distinct form of social information-aggregated opinion-increasingly guides our judgments and decisions. We investigated how people learn from such information by conducting three experiments with participants recruited online within the United States (N = 886) comparing the predictions of three computational models: a Bayesian solution to this problem that can be implemented by a simple strategy for combining proportions with prior beliefs, and two alternatives from epistemology and economics. Across all studies, we found the strongest concordance between participants' judgments and the predictions of the Bayesian model, though some participants' judgments were better captured by alternative strategies. These findings lay the groundwork for future research and show that people draw systematic inferences from aggregated opinion, often in line with a Bayesian solution.

Interpretable machine learning model based on clinical factors for predicting muscle radiodensity loss after treatment in ovarian cancer.

Muscle radiodensity loss after surgery and adjuvant chemotherapy is associated with poor outcomes in ovarian cancer. Assessing muscle radiodensity is a real-world clinical challenge owing to the requirement for computed tomography (CT) with consistent protocols and labor-intensive processes. This study aimed to use interpretable machine learning (ML) to predict muscle radiodensity loss. This study included 723 patients with ovarian cancer who underwent primary debulking surgery and platinum-based chemotherapy between 2010 and 2019 at two tertiary centers (579 in cohort 1 and 144 in cohort 2). Muscle radiodensity was assessed from pre- and post-treatment CT acquired with consistent protocols, and a decrease in radiodensity ≥ 5% was defined as loss. Six ML models were trained, and their performances were evaluated using the area under the curve (AUC) and F1-score. The SHapley Additive exPlanations (SHAP) method was applied to interpret the ML models. The CatBoost model achieved the highest AUC of 0.871 (95% confidence interval, 0.870-0.874) and F1-score of 0.688 (95% confidence interval, 0.685-0.691) among the models in the training set and outperformed in the external validation set, with an AUC of 0.839 and F1-score of 0.673. Albumin change, ascites, and residual disease were the most important features associated with a higher likelihood of muscle radiodensity loss. The SHAP force plot provided an individualized interpretation of model predictions. An interpretable ML model can assist clinicians in identifying ovarian cancer patients at risk of muscle radiodensity loss after treatment and understanding the contributors of muscle radiodensity loss.

Machine learning for automating subjective clinical assessment of gait impairment in people with acquired brain injury – a comparison of an image extraction and classification system to expert scoring

BackgroundWalking impairment is a common disability post acquired brain injury (ABI), with visually evident arm movement abnormality identified as negatively impacting a multitude of psychological factors. The International Classification of Functioning, Disability and Health (ICF) qualifiers scale has been used to subjectively assess arm movement abnormality, showing strong intra-rater and test-retest reliability, however, only moderate inter-rater reliability. This impacts clinical utility, limiting its use as a measurement tool. To both automate the analysis and overcome these errors, the primary aim of this study was to evaluate the ability of a novel two-level machine learning model to assess arm movement abnormality during walking in people with ABI.MethodsFrontal plane gait videos were used to train four networks with 50%, 75%, 90%, and 100% of participants (ABI: n = 42, healthy controls: n = 34) to automatically identify anatomical landmarks using DeepLabCut™ and calculate two-dimensional kinematic joint angles. Assessment scores from three experienced neurorehabilitation clinicians were used with these joint angles to train random forest networks with nested cross-validation to predict assessor scores for all videos. Agreement between unseen participant (i.e. test group participants that were not used to train the model) predictions and each individual assessor’s scores were compared using quadratic weighted kappa. One sample t-tests (to determine over/underprediction against clinician ratings) and one-way ANOVA (to determine differences between networks) were applied to the four networks.ResultsThe machine learning predictions have similar agreement to experienced human assessors, with no statistically significant (p < 0.05) difference for any match contingency. There was no statistically significant difference between the predictions from the four networks (F = 0.119; p = 0.949). The four networks did however under-predict scores with small effect sizes (p range = 0.007 to 0.040; Cohen’s d range = 0.156 to 0.217).ConclusionsThis study demonstrated that machine learning can perform similarly to experienced clinicians when subjectively assessing arm movement abnormality in people with ABI. The relatively small sample size may have resulted in under-prediction of some scores, albeit with small effect sizes. Studies with larger sample sizes that objectively and automatically assess dynamic movement in both local and telerehabilitation assessments, for example using smartphones and edge-based machine learning, to reduce measurement error and healthcare access inequality are needed.

Forecasting the Risk of Cryptocurrencies: Comparison and Combination of GARCH and Stochastic Volatility Models

Abstract The high returns of cryptocurrencies have attracted many investors in recent years. At the same time the evolution of cryptocurrencies is characterized by extreme volatility. For investors, it is therefore key to gauge the risks related to an investment in cryptocurrencies. We provide a comparison of several GARCH and stochastic volatility models for forecasting the risk of cryptocurrencies over the out-of-sample period from 28.09.2018 to 28.02.2023. It turns out that the widely used GARCH(1,1) does not provide accurate risk predictions. In contrast, adding t-distributed innovations or allowing for regime changes improves the accuracy in both model classes. Finally, we consider a Bayesian decision-guided approach with discount learning to combine the different models and provide robust evidence that combining the model predictions leads to accurate combined risk predictions.

The structure and mechanics of the cell cortex depend on the location and adhesion state

Cells exist in different phenotypes and can transition between them. A phenotype may be characterized by many different aspects. Here, we focus on the example of whether the cell is adhered or suspended and choose particular parameters related to the structure and mechanics of the actin cortex. The cortex is essential to cell mechanics, morphology, and function, such as for adhesion, migration, and division of animal cells. To predict and control cellular functions and prevent malfunctioning, it is necessary to understand the actin cortex. The structure of the cortex governs cell mechanics; however, the relationship between the architecture and mechanics of the cortex is not yet well enough understood to be able to predict one from the other. Therefore, we quantitatively measured structural and mechanical cortex parameters, including cortical thickness, cortex mesh size, actin bundling, and cortex stiffness. These measurements required developing a combination of measurement techniques in scanning electron, expansion, confocal, and atomic force microscopy. We found that the structure and mechanics of the cortex of cells in interphase are different depending on whether the cell is suspended or adhered. We deduced general correlations between structural and mechanical properties and show how these findings can be explained within the framework of semiflexible polymer network theory. We tested the model predictions by perturbing the properties of the actin within the cortex using compounds. Our work provides an important step toward predictions of cell mechanics from cortical structures and suggests how cortex remodeling between different phenotypes impacts the mechanical properties of cells.

3-D curved composite façade elements with PV: Results of a pilot project

This paper describes the design, manufacturing, installation and monitoring of a BIPV system that consists of coloured 3D structured façade elements, mounted on an existing office building at the High Tech Campus in Eindhoven. The façade elements contain thin film CIGS modules, which adopt the curved structure of the elements. The encapsulation of the PV modules into the façade elements, the colouration and the curvature led to a relative loss of about 10 % of the efficiency. The PV façade consists of 48 elements with a STC capacity of 5.8 kWp. Each element was equipped with a power optimizer, enabling individual monitoring of each façade element. The elements have been installed at the east and south façade of the building. The monitoring presented in this paper comprised a period of 5 months: the summer and autumn of 2023. The DC performance ratio during that period was 75 % for the elements on the East façade and 80 % for the elements on the South façade; which is in accordance with simplified model predictions. The result is a visual attractive retrofitted PV façade contributing to zero emission strategy of the Eindhoven High Tech Campus.

Short-term solar irradiance forecasting under data transmission constraints

We report a data-parsimonious machine learning model for short-term forecasting of solar irradiance. The model follows the convolutional neural network – long-short term memory architecture. Its inputs include sky camera images that are reduced to scalar features to meet data transmission constraints. The model focuses on predicting the deviation of irradiance from the persistence of cloudiness (POC) model. Inspired by control theory, a noise signal input is used to capture the presence unknown and/or unmeasured input variables and is shown to improve model predictions, often considerably. Five years of data from the NREL Solar Radiation Research Laboratory were used to create three rolling train-validate sets and determine the best representations for time, the optimal span of input measurements, and the most impactful model input data (features). For the chosen test data, the model achieves a mean absolute error of 74.29 W/m2over a time horizon of up to two hours, compared to a baseline 134.35 W/m2 using the POC model.

Construction and application of fetal loss risk model in systemic lupus erythematosus patients with mild disease severity

BackgroundThis dynamic nomogram model was developed to predict the probability of fetal loss in pregnant patients with systemic lupus erythematosus (SLE) with mild disease severity before conception.MethodsAn analysis was conducted on 314 pregnancy records of patients with SLE who were hospitalized between January 2015 and January 2022 at Shenzhen People's Hospital, and the Longhua Branch of Shenzhen People's Hospital. Data from the Longhua Branch of the Shenzhen People's Hospital were utilized as an independent external validation cohort. The nomogram, a widely used statistical visualization tool to predict disease onset, progression, prognosis, and survival, was created after feature selection using multivariate logistic regression analysis. To evaluate the model prediction performance, we employed the receiver operating characteristic curve, calibration curve, and decision curve analysis.ResultsLupus nephritis, complement 3, immunoglobulin G, serum albumin, C-reactive protein, and hydroxychloroquine were all included in the nomogram model. The model demonstrated good calibration and discriminatory power, with an area under the curve of 0.867 (95% confidence interval: 0.787–0.947). According to decision curve analysis, the nomogram model exhibited clinical importance when the probability of fetal loss in patients with SLE ranged between 10 and 70%. The predictive ability of the model was demonstrated through external validation.ConclusionThe predictive nomogram approach may facilitate precise management of pregnant patients with SLE with mild disease severity before conception.

Enhanced lake elevation mapping using a zone-based method

Abstract Inland lakes play a crucial role in monitoring global climate change and managing responses to extreme weather events, with lake elevation being critical for assessing their regulatory capacities. However, due to the limited temporal resolution of current altimetry satellites, obtaining high-frequency, high-precision elevation data for water bodies remains challenging. Consequently, most studies utilize elevation-area (E-A) models constructed from historical elevation and area records, integrated with area observations from high-temporal resolution optical satellites to infer precise water levels. Yet, the construction of the E-A model often assumes a uniform water level across the lake, thus overlooking potential segmentation during dry periods. To address this, our study implemented a zone-based approach, utilizing hydrological connectivity principles to ensure that elevation data within E-A models are confined to appropriate zonal regions. This method effectively minimized uncertainties by preventing errors from zonal discrepancies, significantly improving accuracy compared to traditional methods. It reduced root mean square errors (RMSE) by 0.71 to 1.73 m during the dry season, achieving RMSEs of 0.35, 0.64, and 0.37m across three segments. Furthermore, this method ensures water level data are confined to specific zones, preventing the inconsistencies typically caused by averaging data across multiple stations or selecting data from varying elevations. This consistent domain definition reduces extrapolation errors during the model prediction and inversion. Moreover, by synchronizing data expansion with temporal points, the method compensates for time information losses often incurred by relying on multi-year per-centile charts, thereby enabling more precise aquatic boundary delineation than traditional regional boundaries.

Analysis on cause of erosion of guide vane of high‐head Francis turbine in sandy river

AbstractThe Kizilsu River Basin in China is located in the desert Gobi area with high sediment content and high hardness. After 4798h operation, the maintenance of a high‐head Francis turbine in this basin found that the guide vane was seriously worn, and even most of the top and lower skirts of the front of the guide vane were worn off. In this study, the k‐ε turbulence model, ZGB cavitation model and sediment erosion prediction model were used to simulate the solid‐liquid two‐phase flow and cavitation of guide vane. The research results show that the guide vane of a high‐head Francis turbine in a sandy river has a high sediment velocity and serious surface erosion, with the maximum erosion rate of 1.0 × 10−6 kg/(m2·s), while the area near the skirt of the guide vane is a low‐pressure area lower than the saturated steam pressure, and cavitation is very easy to occur, with the maximum vapor volume fraction of 0.9. Serious cavitation erosion occurs near the skirt of the guide vane, and the combined effect of sand erosion and cavitation erosion on the surface of the skirt of the guide vane aggravates the damage of the skirt structure. The research results provide a technical basis for the antiabrasion design of the guide mechanism of the high‐head Francis turbine and the operation and maintenance of the hydropower station in the sandy river.

Influence of Western Pacific Madden–Julian Oscillation on New York City's Record‐Breaking Air Pollution in Early June 2023

AbstractIn early June 2023, New York City (NYC) and other cities in the northeastern US experienced a severe air pollution event. Although reports associated this hazardous pollution event with the smoke from Canadian wildfires, the factors triggering the southward waft of the smoke remain unclear. We found the northerly anomaly that transported the smoke was linked to the Rossby wave train excited by the Madden–Julian Oscillation (MJO) over the Philippine Sea, which led to the formation of an enhanced northerly at the western edge of the cyclonic anomaly over the East Coast–North Atlantic. When the MJO convection left the western Pacific, the disorganized teleconnection caused the pollution to dissipate. Observational findings were further supported by model simulations and predictions. These results suggest that monitoring and predictions of MJO activity may help mitigate air pollution events over the northeastern US during Canadian wildfire seasons.

Quantifying the mineral magnetic signature of petroleum systems and their source rocks: A study on the Inner Moray Firth, UK North Sea

Summary This study aims to expand on existing connections between magnetic minerals and hydrocarbons within petroleum systems. Previous studies have focussed on single-source petroleum systems whereas this study, for the first time, analyses a multi-source petroleum system to investigate potential correlations between different kerogen type source rocks and magnetic minerals. To do this, the study investigates the magnetic mineral characteristics of the Inner Moray Firth (IMF), UK North Sea, through room-, low-, and high-temperature techniques, and correlates this to published basin and petroleum systems modelling results that show a three-source hydrocarbon mix. Magnetic mineral analysis identifies extensive evidence for magnetite, goethite, and siderite, alongside more minor lepidocrocite and iron sulphides. Although we find that magnetite is ubiquitous within the IMF, its abundance is relatively low, and, in contrast, the relatively magnetically weak goethite is more likely the most abundant magnetic mineral throughout the IMF. In agreement with previous studies, we find magnetic enhancement at oil-water contacts (OWCs); however, here, we identify two different magnetic enhancement processes at OWCs in wells, which are dependent on the amount of sulphur available in the local environment. Wells with low levels of sulphur have increasing levels of magnetite towards the OWC, with the magnetic enhancement occurring at the top of the water-saturated section. Sulphur-rich environments display an increase in iron sulphides near the OWC at the bottom of the oil-saturated sediments. Additionally, we confirm the presence of siderite as indicator of upward vertical migration. Combining with petroleum system model predictions, we find direct links between iron hydroxide presence and Type I and II-III kerogen source rocks, and iron sulphide presence with Type II kerogen source rocks. This study shows the potential for further utilisation of magnetic mineral analysis within hydrocarbon exploration and petroleum system definition.

Tensions in e+e-→π+π-(γ)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$e^+e^-\\rightarrow \\pi ^+\\pi ^-(\\gamma )$$\\end{document} measurements: the new landscape of data-driven hadronic vacuum polarization predictions for the muon g-2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$g - 2$$\\end{document}

The situation of the experimental data used in the dispersive evaluation of the hadronic vacuum polarization contribution to the anomalous magnetic moment of the muon is assessed in view of two recent measurements: e+e-→π+π-\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$e^+e^- \\rightarrow \\pi ^+\\pi ^-$$\\end{document} cross sections in the ρ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\rho $$\\end{document} resonance region by CMD-3 and a study of higher-order radiative effects in the initial-state-radiation processes e+e-→μ+μ-γ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$e^+e^- \\rightarrow \\mu ^+\\mu ^-\\gamma $$\\end{document} and e+e-→π+π-γ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$e^+e^- \\rightarrow \\pi ^+\\pi ^-\\gamma $$\\end{document} by BABAR. The impact of the latter study on the KLOE and BESIII cross-section measurements is evaluated and found to be indicative of larger systematic effects than uncertainties assigned. The new situation also warrants a reappraisal of the independent information provided by hadronic τ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ au $$\\end{document} decays, including state-of-the-art isospin-breaking corrections. The findings cast a new light on the longstanding deviation between the muon g – 2 measurement and the Standard Model prediction using the data-driven dispersive approach, and the comparison with lattice QCD calculations.

The changes prediction on terrestrial water storage in typical regions of China based on neural networks and satellite gravity data

Accurate prediction of regional terrestrial water storage change (TWSA) is of great significance for water resources planning and management, and early warning of extreme climate disasters. Aiming at the problem that the conventional methods on prediction of TWSA time series are difficult to be accurate, the six typical regions are selected in China as examples, including the upper reaches of the Yangtze River (UYR), the southwest region (SWR), the Liaohe River Basin (LRB), the North China Plain (NCP), the Qinghai-Tibet Plateau (QTP), and the Pearl River Basin (PRB). The mascon product from GRACE/GRACE-FO provided by CSR is used to extract TWSA time series in six typical areas. The improved Back Propagation (BP) neural network, Long Short-Term Memory (LSTM) neural network and the latest Bidirectional LSTM (BiLSTM-attention) neural network model based on attention mechanism are proposed to predict and analyze the regional TWSA. In the experiment, the selection of the optimal model parameters such as the number of hidden layer nodes and the number of hidden units of the neural network model is tested and analyzed in detail. Meanwhile, the model prediction results are compared with the traditional least squares method and random forest (RF) prediction method. The root mean square error (RMSE), determination coefficient (R2), Nash–Sutcliffe efficiency coefficient (NSE) and mean absolute percentage error (MAPE) were used to evaluate the accuracy of the predicted results. The results show that the improved BP, LSTM and Bi-LSTM-attention neural network models all achieve higher prediction accuracy in UYR and SWR areas. RMSE is less than 2.641 cm, R2 is as high as 0.8 or more, NSE is above 0.6, and MAPE is within 0.1. Compared with the least square method, the RMSE of the predicted results from three neural network decreased by 0.998 cm, 0.700 cm and 0.7563 on average, and the R2 increased by 81.75%, 69.89% and 72% on average. Compared with RFML method, the RMSE from three neural network is reduced by 0.601 cm, 0.316 cm and 0.360, and R2 is increased by 38.20%, 24.60% and 27.06% on average. NSE and RMSE are improved to varying degrees in the above regions. It shows that the improved BP, LSTM and BiLSTM-attention model used can effectively predict TWSA. The research methods and results in this paper can provide important reference for the rational utilization of regional water resources and disaster risk assessment.