Learning Framework Research Articles

Motion Hologram: Jointly optimized hologram generation and motion planning for photorealistic 3D displays via reinforcement learning.

Holography is capable of rendering three-dimensional scenes with full-depth control and delivering transformative experiences across numerous domains, including virtual and augmented reality, education, and communication. However, traditional holography presents 3D scenes with unnatural defocus and severe speckles due to the limited space bandwidth product of the spatial light modulator (SLM). Here, we introduce Motion Hologram, a holographic technique that accurately portrays photorealistic and speckle-free 3D scenes. This technique leverages a single hologram and a learnable motion trajectory, which are jointly optimized within a deep reinforcement learning framework. Specifically, we experimentally demonstrated that the proposed technique could achieve a 4- to 5-dB PSNR improvement of focal stacks in comparison with traditional holography and could successfully depict speckle-free, high-fidelity, and full-color 3D displays using only a commercial SLM. We believe that the proposed method promises a prospective form of holographic displays that will offer immersive viewing experiences for audiences.

KubePipe: a container-based high-level parallelization tool for scalable machine learning pipelines

As the complexity and scale of machine learning applications continue to grow, the need for efficient training methodologies becomes increasingly critical. Traditional training processes can be time-intensive, often limiting rapid development and deployment. In response to this challenge, we present KubePipe, a high-level tool that abstracts parallelism and containerization from the user, allowing non-expert users to leverage advanced parallel architectures without requiring deep knowledge of parallel computing or container orchestration. KubePipe enables the concurrent execution of multiple machine learning workflows within a Kubernetes cluster, optimizing computational resources and significantly reducing training times. By leveraging containerized environments, KubePipe ensures a high degree of modularity, scalability, and portability, making it adaptable to various machine learning frameworks and tasks. Our experimental results demonstrate substantial performance improvements when using KubePipe compared to conventional pipeline implementations. This paper explores the architecture and functionality of KubePipe, providing insights into its integration with existing machine learning systems and highlighting its potential to streamline the training process in high-performance computing environments.

Machine Learning Framework with Wilson Loop Perceptron for Measuring the Rate of Mutation Patterns between DNA and RNA Viruses

Machine Learning Framework with Wilson Loop Perceptron for Measuring the Rate of Mutation Patterns between DNA and RNA Viruses

A New Control Approach for a Multi-Controlled Wheelchair Utilizing Deep Learning Framework and Raspberry Pi

This article describes the construction and control mechanisms of a multiple control wheelchair (MCW) aimed to aid individuals with amyotrophic lateral sclerosis (ALS) and lower limb disabilities. The MCW has been tested in the laboratory on a total of 10 patients, among them 3 with lower limb disabilities, 6 elderly patients, and 1 patient with ALS. The proposed wheelchair can navigate in any direction using object detection techniques powered by OpenCV. Employing a deep learning algorithm, it performs real-time object detection and tracks its current position in various environmental conditions. The MCW can be controlled using a joystick, internet of things (IoT), or a combination of both. A novel aspect of this work is the implementation of a deep learning framework on a Raspberry Pi processor, allowing the wheelchair to navigate in any direction based on the user’s needs. The Raspberry Pi interfaces with all sensors, camera, and additional peripherals through serial communication, enabling monitoring and control via an android mobile application. This setup effectively reduces the physical distance between the user and the caregiver. During the test, a range of functional measurements were taken, including assessments of the object detection process and the performance of hardware components such as sensors and motors connected to the system. The current system is unable to precisely determine an object’s position during detection, but ongoing research is focused on enhancing the system’s capability to accurately map objects in 3D space for improved wheelchair navigation.

Machine learning-enhanced high-resolution exposure assessment of ultrafine particles

Ultrafine particles (UFPs) under 100 nm pose significant health risks inadequately addressed by traditional mass-based metrics. The WHO emphasizes particle number concentration (PNC) for assessing UFP exposure, but large-scale evaluations remain scarce. In this study, we developed a stacking-based machine learning framework integrating data-driven and physical-chemical models for a national-scale UFP exposure assessment at 1 km spatial and 1-hour temporal resolutions, leveraging long-term standardized PNC measurements in Switzerland. Approximately 20% (1.7 million) of the Swiss population experiences high UFP exposure exceeding an annual mean of 104 particles‧cm−3, with a national average of (9.3 ± 4.7)×103 particles‧cm−3, ranging from (5.5 ± 2.3)×103 (rural) to (1.4 ± 0.5)×104 particles‧cm−3 (urban). A nonlinear relationship is identified between the WHO-recommended 1-hour and 24-hour exposure reference levels, suggesting their non-interchangeability. UFP spatial heterogeneity, quantified by coefficient of variation, ranges from 4.7 ± 4.2 (urban) to 13.8 ± 15.1 (rural) times greater than PM2.5. These findings provide crucial insights for the development of future UFP standards.

Comprehensive Deep Learning Framework of Image, Audio and Video Forgery Detection

Comprehensive Deep Learning Framework of Image, Audio and Video Forgery Detection

Supporting english speaking practice in higher education: the impact of AI chatbot-integrated mobile-assisted blended learning framework

Supporting english speaking practice in higher education: the impact of AI chatbot-integrated mobile-assisted blended learning framework

Optimizing Disease Detection: A Multi-Modal Deep Learning Framework for Medical Imaging and Clinical Data Integration

This study introduces a novel multi-modal deep learning framework that integrates medical imaging data with clinical records for enhanced disease detection. We propose a hybrid architecture combining convolutional neural networks (CNNs) for image analysis and transformer networks for processing clinical data. The framework was evaluated on a dataset of 10,000 patients over 12 months, focusing on detecting early signs of lung cancer and coronary artery disease. Results show our integrated approach achieves significantly higher accuracy compared to single-modality models, with an F1 score of 0.89 (95% CI: 0.87-0.91, p < 0.001). We also introduce a novel interpretability metric for multi-modal models and demonstrate a 30% improvement in model explainability. These findings suggest our approach can enhance diagnostic accuracy while maintaining interpretability in clinical settings.

XGeoS-AI: an interpretable learning framework for deciphering geoscience image segmentation

XGeoS-AI: an interpretable learning framework for deciphering geoscience image segmentation

Interpretable Deep Learning Framework for COVID-19 Detection: Grad-CAM Integration with Pre-trained CNN Models on Chest X-Ray Images

This study is present a novel approach for interpretability enhancing of the deep learning models (EfficientNet, ResNet, VGG) that applied to COVID-19 diagnosis by using the Gradient-Weighted Class Activation Mapping (Grad-CAM) all that to make transparent decision-making improved. To do this we leveraging the capabilities of Grad-CAM, and we aim to provide not only accurate diagnostic predictions but also give a visual explanations, that support the professionals in the healthcare to understanding the underlying features that aided to the model’s decisions. This interpretability is important for building trust in the AI systems, especially in medical areas diagnosis that critical such. This interpretability is essential for building trust in the AI systems, especially in critical areas such as medical diagnosis, that is allowing healthcare professionals to understand the rationale behind the AI-generated recommendations and decisions. In the context of COVID-19, using techniques like Gradient-Weighted Class Activation Mapping (Grad-CAM) can provide insights into which features of medical imaging data contribute most significantly to the model’s predictions, this enhancing reliability and transparency of the AI system. This capability not only aids clinicians in understanding the rationale behind AI-driven diagnoses but it is also fosters greater trust in the automated systems, especially in high-stakes scenarios like healthcare. It is crucial this transparency is ensuring that healthcare professionals can make informed decisions based on the AI’s outputs. As the COVID-19 pandemic demonstrated, timely and accurate diagnosis is an essential for the effective patient management.

Values in education: animating the care and support for teaching and learning framework

ABSTRACT Inclusive approaches to leadership in schools rely on values to build a shared culture. Within the Care and Support for Teaching and Learning school community, values help us focus our attention on what we consider most important when making choices. They inform our identity and our vision of the future, and the social, political and economic structures we develop and maintain to ensure that vision. This paper attempts to highlight the thread of the national conversation about values. The 2001 Manifesto on Values, Education and Democracy for South African public educationremains the most fundamental expression of these values: Democracy; Social Justice and Equity; Equality; Non-Racism and Non-Sexism; Ubuntu (Human Dignity); An Open Society; Accountability (Responsibility); The Rule of Law; Respect; and Reconciliation. While carefully articulated these ideals have not always found deliberate reflection and expression in everyday leadership of school culture. Furthermore, these values were identified in the context of a transition from apartheid to democracy. We can now ask whether there is a need to identify additional values that go beyond transition and increase the chances of thriving for all within the context of the framework of Care and Support for Teaching and Learning.

Semi-supervised learning-based identification of the attachment between sludge and microparticles in wastewater treatment.

Monitoring the microparticle transfer process in wastewater treatment systems is crucial for improving treatment performance. Supervised deep learning methods show high performance to automatically detect particles, but they rely on vast amounts of labeled data for training. To overcome this issue, we proposed a semi-supervised learning (SSL) method based on the Simple framework for Contrastive Learning of visual Representations (SimCLR), to detect microparticles free from sludge and attached to sludge. First, we pre-trained a ResNet50 backbone by SimCLR, to extract features from much unlabeled data (1,000 images). Then, we constructed a Mask R-CNN architecture based on the pre-trained ResNet50, and fine-tuned it on a small quantity of labeled data (≈200 images with ≈600 annotated particles) in supervised learning fashion. We showcased its performance and practical applicability for microscopy images obtained from the water lab of TU Delft. The results demonstrate that the SSL methods obtain a significant improvement in mean average precision of up to 5% compared to the conventional supervised learning method, when a limited amount of labeled data is available (e.g., 91 labeled images). Furthermore, these methods improve the average precision for detecting attached particles by over 12%. With the detection results from the SSL methods, we measured the attachment efficiency of microparticles to sludge under varying mixed liquor suspended solids concentration and aeration intensity. The precise measurements demonstrate the effectiveness and practical applicability of the SSL method in facilitating long-term monitoring of particle transfer processes in biological wastewater treatment systems.

Advancing public safety and housing solutions: A comprehensive framework for machine learning and predictive analytics in urban policy optimization

Urban areas worldwide face mounting challenges, including public safety concerns, housing shortages, and efficient resource allocation. Addressing these complexities requires innovative solutions that integrate technology and policy frameworks. This paper proposes an integrated framework leveraging machine learning (ML) and predictive analytics to optimize urban policy-making and provide actionable insights for sustainable urban development. The framework combines real-time data streams with advanced predictive modeling to enhance decision-making in urban governance. Key features include the ability to assess and respond to emerging trends in public safety by analyzing crime patterns, resource deployment, and community risk factors. Similarly, the framework addresses housing shortages by evaluating demand-supply dynamics, forecasting population growth, and identifying areas requiring immediate intervention. A critical review of existing models underscores the limitations of traditional urban policy approaches in managing socioeconomic disparities and environmental influences. Building upon these insights, the proposed framework integrates diverse data sources, such as demographic information, environmental metrics, and public sentiment analysis, to create a holistic view of urban systems. This multidimensional approach ensures that policy recommendations are context-sensitive, equitable, and inclusive. Moreover, the framework emphasizes stakeholder engagement, fostering collaboration among policymakers, urban planners, community leaders, and data scientists. It prioritizes scalability and adaptability, allowing its application across diverse urban contexts and geographic locations. This adaptability ensures that the framework remains relevant as urban environments evolve and new challenges arise. By addressing the interplay between data-driven insights and policy optimization, this research aims to bridge the gap between theoretical models and practical implementation. The framework holds the potential to redefine urban governance by enabling proactive decision-making, equitable resource allocation, and sustainable development. Future research should explore real-world applications of this framework to validate its efficacy and refine its components further.

How machine learning can revolutionize building comfort: Accessing the impact of occupancy prediction models on HVAC control system

Ventilation control systems globally constitute among the most significant energy demands in the building industry. Optimizing the energy usage of such systems is imperative for constructing sustainable buildings and is essential for achieving environmental sustainability. Occupancy factors of these buildings, particularly Heating, Ventilation and Air Conditioning (HVAC) optimization, are pivotal for energy optimization. In this work, we apply machine learning approaches to improve the efficiency of the HVAC systems of the Engineering Classroom and Research Building (ENCARB) at Prairie View A&M University. We focus on supporting real-time automated HVAC control through the accurate estimation of HVAC temperature based on occupancy patterns. Therefore, we introduce AIRFLO, an AI-powered Robust Framework for Learning to Optimize HVAC energy consumption. Our framework integrates several occupancy factors obtained from the class schedule to estimate ideal HVAC temperatures. Specifically, we incorporate user nonspecific behavior vis-a-vis energy HVAC service period within the ENCARB Building to gather useful matrices as the basis for the model design. To learn the complex patterns in data, we trained our framework using supervised machine learning. Specifically, we initially trained our framework using an ensemble of multilayer neural networks using our training data and observed the estimation performance on an independent validation set. To learn deeper representations and perform systematic comparative model analysis, we proposed our plan to incorporate both Convolutional Neural Networks (CNNs) and Graph Neural Networks (GNNs). Overall, our proposed approach will offer a comprehensive and scalable framework for optimizing HVAC energy optimization, leading to improved sustainability.

Causality enhanced deep learning framework for quality characteristic prediction via long sequence multivariate time-series data

Causality enhanced deep learning framework for quality characteristic prediction via long sequence multivariate time-series data

Epigenetic Impacts of Non-Coding Mutations Deciphered Through Pre-Trained DNA Language Model at Single-Cell Resolution.

DNA methylation plays a critical role in gene regulation, affecting cellular differentiation and disease progression, particularly in non-coding regions. However, predicting the epigenetic consequences of non-coding mutations at single-cell resolution remains a challenge. Existing tools have limited prediction capacity and struggle to capture dynamic, cell-type-specific regulatory changes that are crucial for understanding disease mechanisms. Here, Methven, a deep learning framework designed is presented to predict the effects of non-coding mutations on DNA methylation at single-cell resolution. Methven integrates DNA sequence with single-cell ATAC-seq data and models SNP-CpG interactions over 100kbp genomic distances. By using a divide-and-conquer approach, Methven accurately predicts both short- and long-range regulatory interactions and leverages the pre-trained DNA language model for enhanced precision in classification and regression tasks. Methven outperforms existing methods and demonstrates robust generalizability to monocyte datasets. Importantly, it identifies CpG sites associated with rheumatoid arthritis, revealing key pathways involved in immune regulation and disease progression. Methven's ability to detect progressive epigenetic changes provides crucial insights into gene regulation in complex diseases. These findings demonstrate Methven's potential as a powerful tool for basic research and clinical applications, advancing this understanding of non-coding mutations and their role in disease, while offering new opportunities for personalized medicine.

A Model for Diagnosing Mild Nutrient Stress in Facility-Grown Tomatoes Throughout the Entire Growth Cycle

The effective diagnosis of mild nutrient stress across the complete growth cycle of facility-grown tomatoes is challenging. This study proposes a deep learning framework based on CNN + LSTM, using canopy near-infrared spectroscopy from different growth stages of tomatoes as input, to diagnose mild stress of nitrogen (N), potassium (K), and calcium (Ca) throughout the entire growth cycle of facility-grown tomatoes. The study compares the diagnostic performance of Random Forest (RF), Support Vector Machine (SVM), Partial Least Squares (PLS), Convolutional Neural Networks (CNNs), and CNN + Long Short-Term Memory (LSTM) models for detecting mild nutrient stress in facility-grown tomatoes. Firstly, the preprocessing method of spectral characteristic bands combined with Savitzky‒Golay (SG) + Standard Normal Variate (SNV) was determined. Subsequently, all sample data were divided into six groups: N-deficient, K-deficient, Ca-deficient, N-excess, K-excess, and Ca-excess. The aforementioned models were then used for classification prediction. The results show that RF and CNN + LSTM models demonstrated good predictive performance. Specifically, RF achieved accuracy rates of 70.14%, 90.81%, 88.59%, and 85.37% in the classification tasks of Ca-deficient, N-excess, K-excess, and Ca-excess, respectively. The CNN + LSTM model achieved accuracy rates of 93.33%, 63.33%, 99.2%, 83.33%, and 98.52% in the classification tasks of K-deficient, Ca-deficient, N-excess, K-excess, and Ca-excess, respectively. Finally, in the Leave-One-Group-Out Validation (LOGOV) for validating the model’s generalisation performance, RF performed better in the N-deficient, K-deficient, and Ca-deficient tasks, achieving diagnostic accuracy rates of 80.19%, 81.43%, and 77.02%, respectively. The CNN + LSTM model showed a diagnostic accuracy rate of 66.72% in the N-excess classification task. The study concludes that, given complete training data, the CNN + LSTM model can effectively diagnose mild nutrient stress (N, K, and Ca) in facility-grown tomatoes in most scenarios.

A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates

A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates

ECG signal generation using feature disentanglement auto-encoder

Objective.The demand for electrocardiogram (ECG) datasets, particularly those containing rare classes, poses a significant challenge as deep learning becomes increasingly prevalent in ECG signal research. While generative adversarial networks (GANs) and variational autoencoders (VAEs) are widely adopted, they encounter difficulties in effectively generating samples for classes with limited instances.Approach.To address this issue, we propose a novelFeatureDisentanglement Auto-Encoder (FDAE) designed to dissect various generative factors under a contrastive learning framework within ECG data to facilitate the generation of new ECG samples. The FDAE enhances and extends the AE structure with novel methodologies, which involve: (1) partitioning the latent space into three distinct representations to capture various generative factors; (2) utilizing a contrastive loss function to improve feature disentanglement capabilities; and (3) incorporating additional classifiers to enhance representation learning, alongside a discriminator aimed at boosting the realism of synthesized signals. Furthermore, our FDAE generates new signals by swapping latent codes of existing signals and combining freely or substituting patient-independent representations with those randomly generated by a VAE.Main results.To validate our approach, we conduct heartbeat classification experiments on the publicly available MIT-BIH arrhythmia database, using FAKE-train/FAKE-test partitions and data augmentation. The results highlight the FDAE's ability to improve ECG classifier performance and excel in synthesizing ECG signals. Furthermore, we apply the model to the Icentia11K dataset and conducted classification enhancement experiments. The results further highlight the model's strong generalization ability in ECG synthesis.Significance.This work has the potential to improve the robustness and generalization of deep learning models for ECG analysis, particularly in medical applications where rare cardiac events are often underrepresented in available datasets.

Applying YOLOv6 as an ensemble federated learning framework to classify breast cancer pathology images

The most common carcinoma-related cause of death among women is breast cancer. Early detection is crucial, and the manual screening method may lead to a delayed diagnosis, which would delay treatment and put lives at risk. Mammography imaging is advised for routine screening to diagnose breast cancer at an early stage. To improve generalizability, this study examines the implementation of Federated Learning (FedL) to detect breast cancer. Its performance is compared to a centralized training technique that diagnoses breast cancer. Although FedL has been famous as a safeguarding privacy algorithm, its similarities to ensemble learning methods, such as federated averaging (FEDAvrg), still need to be thoroughly investigated. This study examines explicitly how a YOLOv6 model trained with FedL performs across several clients. A new homomorphic encryption and decryption algorithm is also proposed to retain data privacy. A novel pruned YOLOv6 model with FedL is introduced in this study to differentiate benign and malignant tissues. The model is trained on the breast cancer pathological dataset BreakHis and BUSI. The proposed model achieved a validation accuracy of 98% on BreakHis dataset and 97% on BUSI dataset. The results are compared with the VGG-19, ResNet-50, and InceptionV3 algorithms, showing that the proposed model achieved better results. The tests reveal that federated learning is feasible, as FedAvrg trains models of outstanding quality with only a few communication rounds, as shown by the results on a range of model topologies such as ResNet50, VGG-19, InceptionV3, and the proposed Ensembled FedL YOLOv6.