The demand for food all over the world requires the implementation of advanced technologies to improve agricultural productivity. Federated Learning (FL) as a decentralized approach to machine learning facilitates collaborative model training on different data sources while maintaining privacy-making it highly applicable technology for sensitive agricultural data. This paper offers a systematic overview of the recent knowledge on the application of FL towards the prediction of crop yield and other agricultural uses. We discussed the mathematical basis of FL, the variety of machine learning models used, the types of used agricultural data, and the major performance metrics. The paper presents real-world applications and lists the current limitations, including communication overhead, data heterogeneity, and interpretability issues. Lastly, we introduce open research directions to inform the development of FL in precision agriculture.
This study investigates the effectiveness of various regression models for predicting housing prices using the California Housing dataset. The models evaluated include Linear Regression, Ridge Regression, Best Polynomial Ridge Regression, Lasso Regression, Elastic Net Regression, and Support Vector Regression (SVR) with an RBF kernel. The analysis reveals that SVR with an RBF kernel exhibits the poorest performance, characterized by the highest Mean Squared Error (MSE) and the lowest R² score, indicating limited effectiveness for this dataset. Conversely, Linear Regression, Ridge Regression, and Best Polynomial Ridge Regression demonstrate significantly lower MSE values and nearly identical R² scores, each explaining approximately 60% of the variance in housing prices. Among these, Best Polynomial Ridge Regression marginally outperforms Linear and Ridge Regression, suggesting that including polynomial features enhances model performance. Although Lasso Regression shows slightly higher MSE than the leading models, it still performs better than Elastic Net and SVR. Overall, the study identifies linear regression, ridge regression, and best polynomial ridge regression as the most effective models for this dataset, with SVR and an RBF kernel being the least effective.
Pomegranates are among the many vital crops generally believed to offer health quality and economically impact agriculture. Accurate detection and classification of the pomegranate growth stages enables fruit harvesting robots, resulting in yield optimization, supply chain, and market readiness. In this study, we propose that the YOLO object detection models adopt a novel curriculum learning approach to detect the growth stages of pomegranates using multi-source datasets. A combined dataset of 5700+ images is considered for categorizing pomegranates into bud, flowering, early-fruit, mid-growth, and maturity stages. In the present study, we have experimented with two deep learning-based object detection models, YOLOv5 and YOLOv7, to perform training using a multi-level curriculum learning approach. The proposed training strategy proves to show improvement over state-of-the-art work Zhao et al. [2] in achieving higher detection accuracies towards all five classes. The efficiency of the curriculum learning strategy with YOLOv7 and YOLOv5 models achieved a mAP score of 92.2% with a precision of 90.1%, recall of 82.3%, and F1 score of 86.1%. By comparing the performance proposed curriculum learning-based YOLO object detection models with traditional learning-based YOLO models, it was revealed that the YOLOv5 model with curriculum learning shows a consistent improvement over another model. It is inferred that the proposed training strategy is valuable in increasing the efficacy of pomegranate fruit detection and enhancing precision agriculture activities.
The objective of the study is to enhance uncertainty prediction in regression problems by introducing a revolutionary Bayesian Neural Network (BNN) model. Experimental results reveal significant improvements in uncertainty prediction and point forecasts with the integrated BNN model compared to the plain BNN. Performance metrics, including mean squared error (MSE), mean absolute error (MAE), and R-squared (R²), demonstrate superior results for the proposed BNN. The experimental results show that for plain BNN, MSE is 87.3, MAE is 6.62 and R2 is −0.0492, whereas for the proposed BNN model MSE was found to be 44.64, MAE is 4.4 and R2 is 0.46. This research brings a fresh approach to Bayesian Neural Networks by incorporating both dropout and KL regularization techniques, resulting in a powerful tool for handling regression tasks with certainty. By combining these techniques, the study enhances model stability, avoids overfitting, and achieves more reliable uncertainty estimation. This study adds to our knowledge of uncertainty-aware machine learning models and offers a valuable solution for accurately assessing uncertainty in various applications. • The innovative BNN model merges the power of Bayesian principles with the effectiveness of dropout and KL regularization. • To test and refine our model, the study utilizes the Boston Housing dataset for both training and evaluation purposes.
Over the last few years, the conceptualization of Smart Home has received acceptance. The extensive issues regarding a smart home include offloading computational tasks, data security aspects, privacy issues, authentication of Internet of Things (IoT) devices, and so on. Presently, existing smart home automation addresses either of these issues, nevertheless, Smart Home automation that also necessitates decision-making for offloading computational tasks with improved QoS (i.e., latency and throughput) and systematic features apart from being reliable and safe is a definite necessity. To address these gaps in this, work a QoS-improved method called, Blockchain-modeled Swarm Optimized Lyapunov Smart Contract Deep Reinforced Tasks Offloading (BSOLSC-DRTO) in smart home is proposed. The BSOLSC-DRTO method is split into two sections, namely, Offloading Computational Tasks based on the Particle Swarm Optimized Lyapunov model and Temporal Difference Deep Reinforced Secured Offloading. First to solve the offloading issue and therefore improve the QoS, we developed a Particle Swarm Optimized Lyapunov model using a Lyapunov optimization function. This optimization problem aims to minimize latency and improve throughput considerably. Second, to boost the offloading security, we propose a trustworthy access control using the Temporal Difference Deep Reinforced Secured Offloading model that can safeguard devices against illegal offloading. Then to handle the computation management for addressing the offloading decisions in the queue temporal difference function is applied, therefore improving the smart contract accuracy and precision involved in offloading computational tasks. Evaluation results from experiments and numerical simulations exhibit the notable advantages of the proposed BSOLSC-DRTO method over existing methods.•Develop a Particle Swarm Optimized Lyapunov model to minimize latency and significantly improve throughput.•Proposed a Temporal Difference Deep Reinforced Secured Offloading model for trustworthy access control, protecting devices against illegal offloading.
Stochastic Calculus-guided Reinforcement learning (SCRL) is a new way to make decisions in situations where things are uncertain. It uses mathematical principles to make better choices and improve decision-making in complex situations. SCRL works better than traditional Stochastic Reinforcement Learning (SRL) methods. In tests, SCRL showed that it can adapt and perform well. It was better than the SRL methods. SCRL had a lower dispersion value of 63.49 compared to SRL's 65.96. This means SCRL had less variation in its results. SCRL also had lower risks than SRL in the short- and long-term. SCRL's short-term risk value was 0.64, and its long-term risk value was 0.78. SRL's short-term risk value was much higher at 18.64, and its long-term risk value was 10.41. Lower risk values are better because they mean less chance of something going wrong. Overall, SCRL is a better way to make decisions when things are uncertain. It uses math to make smarter choices and has less risk than other methods. Also, different metrics, viz training rewards, learning progress, and rolling averages between SRL and SCRL, were assessed, and the study found that SCRL outperforms well compared to SRL. This makes SCRL very useful for real-world situations where decisions must be made carefully. • By leveraging mathematical principles derived from stochastic calculus, SCRL offers a robust framework for making informed choices and enhancing performance in complex scenarios. • In comparison to traditional SRL methods, SCRL demonstrates superior adaptability and efficacy, as evidenced by empirical tests.