Prediction Bias Research Articles

A Convolutional Neural Network–Long Short-Term Memory–Attention Solar Photovoltaic Power Prediction–Correction Model Based on the Division of Twenty-Four Solar Terms

The prevalence of extreme weather events gives rise to a significant degree of prediction bias in the forecasting of photovoltaic (PV) power. In order to enhance the precision of forecasting outcomes, this study examines the interrelationships between China’s 24 conventional solar terms and extreme meteorological events. Additionally, it proposes a methodology for estimating the short-term generation of PV power based on the division of solar term time series. Firstly, given that the meteorological data from the same festival is more representative of the climate state at the current prediction moment, the sample data are grouped according to the 24 festival time nodes. Secondly, a convolutional neural network–long short-term memory (CNN-LSTM) PV power prediction model based on an Attention mechanism is proposed. This model extracts temporal change information from nonlinear sample data through LSTM, and a CNN link is added at the front end of LSTM to address the issue of LSTM being unable to obtain the spatial linkage of multiple features. Additionally, an Attention mechanism is incorporated at the back end of the CNN to obtain the feature information of crucial time steps, further reducing the multi-step prediction error. Concurrently, a PV power error prediction model is constructed to rectify the outcomes of the aforementioned prediction model. The examination of the measured data from PV power stations and the comparison and analysis with other prediction models demonstrate that the model presented in this paper can effectively enhance the accuracy of PV power predictions.

Identification of candidate genes and genomic prediction for early heifer pregnancy in Nelore beef cattle

This study aimed to assess the influence of differential weighting in genomic regions harboring candidate causal loci on the prediction accuracy and inflation for early heifer pregnancy (heifers that calved up to 30 months of age) in Nellore (Bos indicus) heifers using the single step genomic BLUP model (ssGBLUP). Phenotypic records of 102,294 Nellore heifers born between 2010 and 2017 were used in this study. The pedigree dataset harbored information from 176,107 animals born between 1998 and 2017, including 5,145 sires and 35,705 dams. Seven different models for genomic prediction were defined by combining the SNP weights obtained in the iterations (1st and 2nd) of the weighted single step GWAS (ssw1GBLUP and ssw2GBLUP) or candidate QTLs reported in the literature. Hence, the lambda (λ) values estimated in the WssGWAS were used to weight the SNPs adjacent to the candidate regions or QTL previously reported in the literature. To estimate the genetic parameters and perform the WssGWAS and WssGBLUP for early heifer pregnancy, a single-trait Bayesian analysis considering a threshold animal model was used. Accuracy, bias, and inflation parameters were evaluated in the validation subset based on the linear regression (LR) method. Genomic windows of ten consecutive SNPs that explained more than 0.5% of the additive genetic variance were selected to explore and determine possible candidate genes. Among the identified genes, we can highlight the PGRMC2, TENM3, GRIP1, TMEM45A, and KLF3, given their roles in endocrine fertility, expression of contractile proteins, average daily gain, dry matter intake, and fat deposition. Several genomic regions associated with QTL related to early heifer pregnancy were identified. The identification of such regions and the respective candidate genes associated with sexual precocity and fertility would contribute to improve the genetic knowledge regarding early sexual precocity of Nellore cattle. The prediction accuracy increased roughly 25.6% using the ssGBLUP compared to BLUP models. The prediction accuracy with the WssGBLUP when incorporating weighted SNPs with the λ values obtained in the 1st (ssw1GBLUP) and 2nd iteration (ssw2GBLUP) of the WssGWAS was higher (∼18%) than that described for the ssGBLUP model. The inflation also increased the weighting of the most relevant SNPs obtained with the GWAS, most likely overestimating the GEBV. The models that weighted SNPs close to QTLs reported in the literature yielded to less biased and deflated predictions compared to ssw1GBLUP and ssw2GBLUP models. Genomic selection is a feasible alternative for genomic evaluation of early heifer pregnancy in Nellore beef cattle by increasing the prediction accuracy of young animals. In addition, the use of information obtained from the WssGWAS is an alternative to increase reliability and reduce genomic prediction bias. Therefore, the results obtained herein indicate that it is possible to improve the prediction accuracy and reduce the bias of genomic prediction by using genomic information and differentially weighted genomic regions harboring candidate QTLs previously reported in the literature for early heifer pregnancy.

Predictors of Trauma-Related Diagnostic Overshadowing Bias

Diagnostic overshadowing occurs when healthcare professionals misattribute an individual’s presenting symptoms to other features of an individual’s clinical presentation. Mental health providers may incorrectly diagnose and treat trauma-exposed individuals due to trauma-related diagnostic overshadowing bias. No research has investigated provider factors associated with this bias. Mental health provider background and training characteristics were examined as predictors of trauma-related diagnostic overshadowing in a sample of mental health providers (N = 210). Hierarchical regression modeling was used to evaluate predictors related to trauma-related diagnostic overshadowing in diagnostic and treatment decision-making. Providers with diagnostic responsibilities in their professional role were significantly less likely to demonstrate bias in some diagnostic and treatment decisions. Older mental health providers and providers with more clients experiencing traumatic stress demonstrated greater bias in some diagnostic responses. Providers with doctoral degrees, compared to all other degree types, demonstrated less bias through some diagnostic and treatment responses. Bias in diagnostic decisions strongly predicted bias in treatment decision-making. Findings suggest that factors related to clinical experience may be important in predicting trauma-related diagnostic overshadowing. Future work should extend this research to understand which factors affect trauma-related diagnostic overshadowing in real-world settings.

Integrating Data Imputation and Augmentation with Interpretable Machine Learning for Efficient Strength Prediction of Fly Ash-Based Alkali-Activated Concretes

Fly ash-based alkali-activated concrete (AAC) is renowned for its superior mechanical performance and sustainability, presenting an attractive alternative to traditional Portland cement concrete. Despite these advantages, the broad compositional range of AACs presents challenges in precisely tailoring material properties. In this context, machine learning (ML) offers promising prospects to streamline and fast-track the development of advanced materials design strategies by predicting mechanical properties from compositional variations. Effective ML model development, however, hinges on the availability of a comprehensive, high-quality dataset. Previous studies often relied on literature-derived datasets, which typically include outliers, noise, and missing values, potentially leading to biased predictions. Moreover, limited dataset sizes could undermine the robustness of the models. Traditional ML methods applied to AACs also tend to lack interpretability. To address these issues, this paper utilizes several data imputation methods and Generative Adversarial Networks (GANs) for data augmentation, effectively doubling the dataset size. Following this, ML algorithms such as Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Neural Networks (NNs) are leveraged to predict compressive strength. The NN model, especially when enhanced by k-nearest neighbors (kNN) imputation (k = 5), demonstrated superior predictive accuracy compared to RF and XGBoost models. Further, SHAP (SHapley Additive exPlanations) analysis reveals key determinants of compressive strength, such as water content, SiO2, and curing conditions. Visualizations such as SHAP violin and river flow plots further elucidated feature contributions and property distributions. Overall, this study provides a robust framework for exploring composition-strength relationships in AACs, advancing the design of these environment-friendly materials.

Wearing myopia glasses on GANs: Mitigating bias for pre-trained Generative Adversarial Networks via online prior perturbation

Pre-trained Generative Adversarial Networks (GANs) can provide rich information and make various downstream tasks beneficial. However, the training process of GANs potentially learns pre-existing biases in the historical data, which are not only inherited, but also amplified to some extent when using the generated data for prediction. Most available methods for bias mitigation utilize pre-processing or in-processing strategies to retrain GANs, which needs to access the raw data and also increases the time and computational cost required to train a fully expressive GAN. Therefore, we propose a novel post-processing mechanism to achieve fairness outputs of GANs through online prior perturbation. Unlike traditional output-based offline processing, our method mitigates biases from model inputs by designing a prior perturbation network, called “prior perturber”, to form a combined network with the pre-trained GAN. Specifically, we introduce a bias prediction network, called “bias predictor”, for online adversarial training with the prior perturber, which can decouple the generated representations from the bias features and improve the fairness of the downstream prediction results. In addition, the training process is independent of specific target labels, and the generated fair representations have good transferability. Experimental evaluations on five real datasets validate the effectiveness of the proposed method and the best utility is obtained on the same fairness level as compared to other output-based baseline methods. The result on fairness-utility trade-offs significantly improves the fairness of original samples by 56.40%, while the utility is reduced by only 2.60%.

Drosophila Eye Gene Regulatory Network Inference Using BioGRNsemble: An Ensemble-of-Ensembles Machine Learning Approach

Background: Gene regulatory networks (GRNs) are complex gene interactions essential for organismal development and stability, and they are crucial for understanding gene-disease links in drug development. Advances in bioinformatics, driven by genomic data and machine learning, have significantly expanded GRN research, enabling deeper insights into these interactions. Methods: This study proposes and demonstrates the potential of BioGRNsemble, a modular and flexible approach for inferring gene regulatory networks from RNA-Seq data. Integrating the GENIE3 and GRNBoost2 algorithms, the BioGRNsemble methodology focuses on providing trimmed-down sub-regulatory networks consisting of transcription and target genes. Results: The methodology was successfully tested on a Drosophila melanogaster Eye gene expression dataset. Our validation analysis using the TFLink online database yielded 3703 verified predicted gene links, out of 534,843 predictions. Conclusion: Although the BioGRNsemble approach presents a promising method for inferring smaller, focused regulatory networks, it encounters challenges related to algorithm sensitivity, prediction bias, validation difficulties, and the potential exclusion of broader regulatory interactions. Improving accuracy and comprehensiveness will require addressing these issues through hyperparameter fine-tuning, the development of alternative scoring mechanisms, and the incorporation of additional validation methods.

Enhancing Machine Learning Models Through PCA, SMOTE-ENN, and Stochastic Weighted Averaging

Predicting survival outcomes in critical accidents has been a focal point in machine learning research. This study addresses several limitations of existing methods, including insufficient management of data imbalance, lack of emphasis on hyperparameter tuning, and proneness to overfitting. Many existing models struggle to generalize effectively on imbalanced datasets or depend on default hyperparameter settings, resulting in biased predictions. By integrating Principal Component Analysis (PCA), hyperparameter optimization, and resampling methods, as well as combining Edited Nearest Neighbors (ENN) with the Synthetic Minority Oversampling Technique (SMOTE), the model significantly improves predictive accuracy and model generalization. An ensemble model combining seven machine learning algorithms—Logistic Regression, Support Vector Machine, KNN, Random Forest, XGBoost, LightGBM, and CatBoost—was applied to predict survival outcomes. Stochastic Weighted Averaging (SWA) was applied to mitigate overfitting and enhance generalization. The accuracy increased from 91.97% to 94.89% after SWA was applied in this specific scenario. The combination of PCA-based dimensionality reduction, hyperparameter tuning, and resampling techniques (ENN + SMOTE) ensured the model handled data imbalance and optimized predictive accuracy. The final model demonstrated excellent performance, with Area Under the Curve (AUC) and Average Precision (AP) values both reaching 0.98, indicating high accuracy and precision. These improvements were validated using the Titanic dataset in a binary classification problem of predicting passenger survival. The results emphasize that ensemble learning, enhanced by SWA, offers a powerful framework for handling imbalanced and complex datasets, providing significant advancements in predictive modeling accuracy. This study provides insights into how machine learning techniques can be effectively combined to solve classification challenges in real-world scenarios.

Data driven cost-sensitive boosted tree for interpretable banking systemic risk prediction

Systemic risk (SR) in the banking sector poses a significant threat to both the financial system and the real economy. Its inherent characteristics of nonlinearity, non-equilibrium, and interconnectedness make it challenging to analyze using conventional statistical methods. In this paper, a cost-sensitive gradient boosting tree algorithm, FLXGBoost, is proposed for predicting SR. FLXGBoost considers the boosted tree, XGBoost as the base framework, boosting trees as the fundamental framework, guaranteeing the robustness of SR prediction. Additionally, to tackle the challenge of extreme data imbalance prevalent in SR prediction tasks, a cost-aware loss function, focal loss, is embedded into the boosted tree to enable FLXGBoost a risk-aware fashion. Moreover, a tree-derived interpretable algorithm SHAP is incorporated into this cost-sensitive solution, making FLXGBoost an accurate and interpretable risk-aware model. Experimental results on a financial risk prediction dataset pertaining to banking SR evince the capacity of FLXGBoost to significantly reduce the misclassification rate of risk banks, thereby mitigating substantial losses attributed to erroneous predictions of risky scenarios. Moreover, compared with classical imbalanced machine learning-based SR prediction approaches, the diverse evaluation metrics of FLXGBoost show that it is a competitive solution for accurate SR prediction. Besides, the explanatory analysis further demonstrates that FLXGBoost is a promising solution to address the issue of biased predictions in imbalanced banking SR in the interpretation perspective.

SGO: An innovative oversampling approach for imbalanced datasets using SVM and genetic algorithms

Imbalanced datasets present a challenging problem in machine learning and artificial intelligence. Since most models typically assume balanced data distributions, imbalanced positive and negative examples can lead to significant bias in prediction or classification tasks. Current over-sampling methods frequently encounter issues like overfitting and boundary bias. A novel imbalanced data augmentation technique called SVM-GA over-sampling (SGO) is proposed in this paper, which integrates Support Vector Machines (SVM) with Genetic Algorithms (GA). Our approach leverages SVM to identify the decision boundary and uses GA to generate new minority samples along this boundary, effectively addressing both over-fitting and boundary biases. It has been experimentally validated that SGO outperforms the traditional methods on most datasets, providing a novel and effective approach to address imbalanced data problems, with potential application prospects and generalization value.

Biparietal diameter vs crown-rump length as standard parameter for late first-trimester pregnancy dating.

To compare the precision of biparietal diameter (BPD) and crown-rump length (CRL) as predictors of gestational age in the human fetus in the late first and early second trimesters, using a population-based approach. We constructed term and gestational-age prediction curves for first-trimester dating, based on 11 041 pregnancies with 12 260 measurements of CRL and/or BPD from a population-based Norwegian clinical database. We used a population-based approach with local linear quantile regression, combined with a time-to-event strategy that compensates for induced births. Term prediction precision was assessed by estimating and comparing the prediction residual curves using a time-to-event analysis. Individual differences in gestational-age predictions from CRL and BPD were assessed using measurements performed on the same fetus on the same day. A sensitivity analysis was performed to evaluate the effect of not distinguishing between non-spontaneous and spontaneous births. CRL and BPD provided almost identical term prediction precision judged from the residual distribution. In about 51% of examinations, the difference in predicted gestational age was 1 day or less; 24% of examinations had a difference of 2 days, 14% had a difference of 3 days, 7% had a difference of 4 days and only 5% of all examinations had a difference of 5 days or more. Incorrectly removing induced births from the analysis, or treating them as spontaneous, would cause a substantial systematic prediction bias of about 2 days. Based on population data, using comparisons at an individual level, our study found that BPD is as precise as CRL when used for first-trimester dating. BPD has advantages from a clinical point of view, since it is technically less challenging and less time-consuming to measure compared with CRL, and can be measured and assessed throughout the entire pregnancy. © 2024 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Convectively Induced Secondary Circulations and Wind‐Driven Heat Fluxes in the Surface Energy Balance Over Land

AbstractIncreased resolution has enabled kilometer‐scale weather and climate models to partially resolve secondary circulations, including horizontal convective rolls (HCRs) and cold pool gust fronts. Although these circulations are ubiquitous in convective boundary layers over land, their impacts on the surface energy balance are largely unknown. Doppler lidar and surface observations were combined with DOE E3SM land model experiments, revealing increased surface winds (5 m/s) and heat fluxes (50 W/m2) in convergent branches of HCRs. Larger wind‐driven flux responses (up to 150 W/m2) were found along gust fronts. Surface energy balance shifts to accommodate wind‐driven fluxes, reducing ground heat conduction and longwave cooling. Our findings from the US Southern Great Plains are broadly relevant to modeling convective boundary layers. In particular, widely used subgrid wind gust parameterizations were found to be physically inconsistent with resolved secondary circulations and could worsen climate prediction biases at kilometer‐scales.

Extreme-value modelling of migratory bird arrival dates: Insights from citizen science data

Abstract Citizen science mobilises many observers and gathers huge datasets but often without strict sampling protocols, resulting in observation biases due to heterogeneous sampling effort, which can lead to biased predictions. We develop a spatiotemporal Bayesian hierarchical model for bias-corrected estimation of arrival dates of the first migratory bird individuals at their breeding sites. Higher sampling effort could be correlated with earlier observed dates. We implement data fusion of two citizen-science datasets with fundamentally different protocols (BBS, eBird) and obtain posterior distributions of the latent process, which contains four spatial components endowed with Gaussian process priors: species niche; sampling effort; position and scale parameters of annual first arrival date. The data layer consists of four response variables: counts of observed eBird locations (Poisson); presence-absence at observed eBird locations (Binomial); BBS occurrence counts (Poisson); first arrival dates (Generalised Extreme-Value). We devise a Markov Chain Monte Carlo scheme and check by simulation that the latent process components are identifiable. We apply our model to several migratory bird species in the northeastern United States for 2001−2021 and find that the sampling effort significantly modulates the observed first arrival dates. We exploit this relationship to effectively bias-correct predictions of the true first arrivals.

Fixing imbalanced binary classification: An asymmetric Bayesian learning approach.

Most statistical and machine learning models used for binary data modeling and classification assume that the data are balanced. However, this assumption can lead to poor predictive performance and bias in parameter estimation when there is an imbalance in the data due to the threshold election for the binary classification. To address this challenge, several authors suggest using asymmetric link functions in binary regression, instead of the traditional symmetric functions such as logit or probit, aiming to highlight characteristics that would help the classification task. Therefore, this study aims to introduce new classification functions based on the Lomax distribution (and its variations; including power and reverse versions). The proposed Bayesian functions have proven asymmetry and were implemented in a Stan program into the R workflow. Additionally, these functions showed promising results in real-world data applications, outperforming classical link functions in terms of metrics. For instance, in the first example, comparing the reverse power double Lomax (RPDLomax) with the logit link showed that, regardless of the data imbalance, the RPDLomax model assigns effectively lower mean posterior predictive probabilities to failure and higher probabilities to success (21.4% and 63.7%, respectively), unlike Logistic regression, which does not clearly distinguish between the mean posterior predictive probabilities for these two classes (36.0% and 39.5% for failure and success, respectively). That is, the proposed asymmetric Lomax approach is a competitive model for differentiating binary data classification in imbalanced tasks against the Logistic approach.

A robust visual question answering approach to reduce multimodal bias

Currently, many visual question answering models have bias problems. Specifically, when the question-answer relationship in the training data shows a more obvious mapping relationship, the model shows poor generalization ability. For such biased predictions, existing research work mainly considers language bias, while ignoring the bias information introduced by images. In order to enhance the robustness of visual question answering models, a bias reduction method is proposed, and on this basis, the influence of language and visual information on bias is explored. Furthermore, two bias learning branches are constructed to capture language bias and the bias caused by language and images respectively, and the bias reduction method is used to obtain more robust prediction results. Finally, according to the difference in prediction probability between the standard visual question answering and bias branches, the samples are dynamically weighted, so that the model can dynamically adjust the learning degree for samples with different bias levels. Experiments on datasets such as VQA-CP v2.0 prove the effectiveness of the proposed method and alleviate the influence of bias on the model.

Ethnicity Detection with Convolutional Neural Network (CNN): Bangladesh Perspectives

Ethnicity detection, or the automatic determination of people's ethnic backgrounds using visual data, has become important in a number of disciplines, including social sciences, demographics, and computer vision. With a focus on Bangladesh's complex cultural landscape, we describe in this paper a novel method for ethnicity detection using convolutional neural networks (CNNs). We gathered a dataset of photographs depicting members of the three main ethnic groups living in the Chittagong hill neighborhood: Chakma, Marma, and Tripura. Utilizing the DenseNet121 model's robust feature extraction skills, we customized the architecture to meet the needs of our particular ethnicity detection application. A wide range of performance indicators were used to train and assess our unique CNN model. The outcomes illustrate accurate ethnicity detection technology's potential for identifying and resolving social inequities while showcasing a promising level of accuracy. To address possible biases in the model's prediction, ethical considerations are also covered. Overall, this work advances knowledge of Bangladesh's ethnic variety and demonstrates the potential of CNNs for ethnicity detection. The results provide up new avenues for investigation and applications that advance a more diverse and inclusive society.

Age estimation at death in individuals over 50 using CT images: An innovative anthropological approach

ObjectivesThe aim of the present work is to develop a novel method for predicting age in individuals over 50 years old, utilizing regression models. MethodsThe conducted study is of an analytical cross-sectional nature, involving a sample of 44 young subjects and 107 elderly subjects. The necessary data for this research were extracted from "The New Mexico Decedent Image Database." Based on the phenomenon of height shrinkage with age, we created models for young subjects and applied them to elderly subjects, allowing us to extract the variables. ResultsWe obtained highly encouraging results with an R2 of 0.73, a mean absolute error of 3.94, and stable cross-validation. We used a Student's t-test, which demonstrated no significant difference between predicted and actual values (p-value >0.05). We also conducted a learning curve analysis and examined residuals against predicted values. This suggests that the forecasts are accurate, with no significant bias in predictions. ConclusionThis work has allowed us to conclude that it is possible to reliably estimate the age of subjects over 50, taking into account age-related physiological and pathological changes.

Turkey’s Hydropower Potential in the Near Future and the Possible Impacts of Climate Change—A Case Study of the Euphrates–Tigris Basin

Hydropower is becoming an important renewable energy source in Turkey, but the ever-changing atmospheric and climatic conditions of Turkey make it very difficult to be projected efficiently. Thus, an efficient estimation technique is crucial for it to be adopted as a reliable energy source in the future. This study evaluates Turkey’s hydropower potential in the Euphrates–Tigris Basin under changing climatic conditions. We adapted an empirical equation to model reservoir outflows, considering the site-specific characteristics of 14 major dams. Initial results from employing a model with a constant empirical coefficient, α, yielded moderate predictive accuracy, with R2 values ranging from 0.289 to 0.612. A polynomial regression identified optimal α values tailored to each dam’s surface area, significantly improving model performance. The adjusted α reduced predictive bias and increased R2 values, enhancing forecast reliability. Seasonal analysis revealed distinct hydropower trends: Ataturk Dam showed a notable decrease of 5.5% in hydropower generation up to 2050, while Birecik and Keban Dams exhibited increases of 2.5% and 2.2%, respectively. By putting these discoveries into practice, water resource management may become more robust and sustainable, which is essential for meeting Turkey’s rising energy needs and preparing for future climatic challenges. This study contributes valuable insights for optimizing reservoir operations, ensuring long-term hydropower sustainability, and enhancing the resilience of water resource management systems globally.

Modulating effects of temperature on CO2-inhibited isoprene emissions in Eucalyptus urophylla

Terrestrial vegetation emits substantial amounts of highly reactive isoprene, significantly impacting atmospheric chemistry and climate change. Both atmospheric carbon dioxide (CO2) concentration and temperature can influence plant isoprene emissions; however, whether these factors have a synergistic effect remains unclear, particularly for tropical/subtropical plants. In this study, we conducted in-situ controlled experiments on Eucalyptus urophylla, a representative tropical/subtropical species, to investigate the seasonal variation in the response of isoprene emissions to CO2 concentrations (ISOP-CO2 response) and to identify potential controlling factors. The results showed that high CO2 exerts a nearly linear inhibitory effect on isoprene emissions, as indicated by the slope of the ISOP-CO2 response curve. This inhibitory effect exhibited evident seasonal changes, with stronger suppression during cooler seasons and weaker suppression during warmer seasons. This finding contrasts with the default ISOP-CO2 response in the MEGAN model, which ignored seasonal variation. Further analysis showed a significant correlation between the slope of the ISOP-CO2 response curve and growth temperature from the past 10 days, indicating that these metrics are effective indicators for predicting seasonal changes. Our findings reveal a synergistic mechanism between temperature and CO2 concentration effects on isoprene emissions. By coupling the effects of growth temperature with the ISOP-CO2 response, this mechanism can be integrated into models to provide more accurate predictions of future isoprene emissions, reducing prediction biases, especially during cooler seasons.

Genomic prediction and validation strategies for reproductive traits in Holstein cattle across different Chinese regions and climatic conditions

Accurate genomic predictions of breeding values for traits included in the selection indexes are paramount for optimizing genetic progress in populations under selection. The size of the reference populations is a major factor influencing the accuracy of genomic predictions, which is even more important for lowly-heritable traits such as fertility and reproduction indicators. Combining data from different geographical regions or countries can be beneficial for genomic prediction of these lowly heritable traits. Therefore, the objectives of this study were to: 1) evaluate the benefits of performing across-regional genomic evaluations for reproduction traits in Chinese Holstein cattle; and, 2) assess the feasibility of validating genomic predictions across environments based on reaction norm models (RNM) and the Linear Regression (LR) method after accounting for genotype-by-environment interactions. Phenotypic records from 194,574 cows collected across 47 farms located in 2 regions of China were used for this study. The reference and validation populations were defined based on birth year for applying the LR validation method. The traits evaluated included: interval from first to last insemination (IFL), conception rate at the first insemination (CR_f), and number of inseminations (NS) recorded in heifers and first-parity cows. The results indicated that combining data from different regions resulted in greater genomic prediction accuracies compared with using data from single regions, with increases ranging from 2.74% to 93.81%. This improvement was particularly notable for the region with the least amount of available data, where the increases ranged from 26.49% to 93.81%. Furthermore, the predictive abilities could be validated for all studied traits based on the LR method across different environments when fitting RNM. The prediction accuracies and bias of genomic breeding values based on RNM were better than regular single-trait animal models in extreme climatic conditions for IFL and NS, whereas limited increases in predictive abilities were observed for CR_f. Across-regional genomic prediction by RNM can account for genotype-by-environment interactions, potentially increase the accuracy of genomic prediction, and predict the performances of individuals in the environments with limited phenotypic data available.

Label-noise learning via uncertainty-aware neighborhood sample selection

Existing deep learning methods often require a large amount of high-quality labeled data. Yet, the presence of noisy labels in the real-world training data seriously affects the generalization ability of the model. Sample selection techniques, the current dominant approach to mitigating the effects of noisy labels on models, use the consistency of sample predictions and observed labels to make clean selections. However, these methods rely heavily on the accuracy of the sample predictions and inevitably suffer when the model predictions are unstable. To address these issues, we propose an uncertainty-aware neighborhood sample selection method. Especially, it calibrates for sample prediction by neighbor prediction and reassigns model attention to the selected samples based on sample uncertainty. By alleviating the influence of prediction bias on sample selection and avoiding the occurrence of prediction bias, our proposed method achieves excellent performance in extensive experiments. In particular, we achieved an average of 5% improvement in asymmetric noise scenarios.