Global Learning Research Articles

MolLoG: A Molecular Level Interpretability Model Bridging Local to Global for Predicting Drug Target Interactions.

Developing new pharmaceuticals is a costly and time-consuming endeavor fraught with significant safety risks. A critical aspect of drug research and disease therapy is discerning the existence of interactions between drugs and proteins. The evolution of deep learning (DL) in computer science has been remarkably aided in this regard in recent years. Yet, two challenges remain: (i) balancing the extraction of profound, local cohesive characteristics while warding off gradient disappearance and (ii) globally representing and understanding the interactions between the drug and target local attributes, which is vital for delivering molecular level insights indispensable to drug development. In response to these challenges, we propose a DL network structure, MolLoG, primarily comprising two modules: local feature encoders (LFE) and global interactive learning (GIL). Within the LFE module, graph convolution networks and leap blocks capture the local features of drug and protein molecules, respectively. The GIL module enables the efficient amalgamation of feature information, facilitating the global learning of feature structural semantics and procuring multihead attention weights for abstract features stemming from two modalities, providing biologically pertinent explanations for black-box results. Finally, predictive outcomes are achieved by decoding the unified representation via a multilayer perceptron. Our experimental analysis reveals that MolLoG outperforms several cutting-edge baselines across four data sets, delivering superior overall performance and providing satisfactory results when elucidating various facets of drug-target interaction predictions.

Taxonomy of purposes, principles, forms, technologies in on-line education: comparative analysis of virtual universities: pedagogical aspects

Importance. The paper presents the findings of comparative pedagogical research done against the backdrop of new tendencies and developments on the global learning landscape. With the pandemic pushing universities online and ensuing digitalization of educational environment with a flurry of multi aspect virtual learning activities, there has emerged a growing need for reliable criteria to assess effectiveness and efficiency in higher education. The research is aimed to build the taxonomy of virtual universities purposes with the focus on the paramount principles, educational forms and technologies they employ.Research Methods. The research methods involved comparative, descriptive, inductive-deductive ones, typical of comparative pedagogy. We studied the purposes, principles, forms and technologies used by the following virtual universities: The UK’s Open University), Canadian Virtual University (CVU), The University of Phoenix (UoP), The Virtual University for Small States of The Commonwealth (VUSSC), The Virtual University of Pakistan.Results and Discussion. The research enabled us to aggregate, analyze and compartmentalize the data on purposes, principles, forms and technologies of different virtual universities, which offers a new viewpoint on i-learning in a new setting. The purposes of virtual universities are presented in the hierarchical order in compliance with the goal-setting levels: expediency, focus (purposes on this level are differentiated according to the type of activity: educational, pedagogical, maintaining and organizational) and goal commitment. We identified a set of basic principles regulating academic activity of virtual universities in the digitalization context of education, described the most popular forms and technologies employed and specified priority activities of virtual universities.Conclusion. The taxonomy of purposes and paramount principles of virtual universities educational activities make these activities unique. The number of factors and variables defining the educational activities of virtual universities is growing, still the ability to choose the learning route which fits a certain individual has always been and will remain important. Tailored learning style (built on individual preferences in the ways of information searching and processing) suggests an opportunity to adjust the situation and learning material to your needs and thus become more efficient in obtaining new information and skills.

MD-GCCF: Multi-view deep graph contrastive learning for collaborative filtering

Collaborative Filtering (CF), a classical recommender system approach, learns users’ interests and behavioral preferences for items through a user–item interaction graph. CF based on graph neural network (GNN) and CF based on graph contrastive learning (GCL) show strong advantages in both modeling multi-layer signals and solving label sparsity, respectively. However, there are still two key problems to be solved: Most CF models based on (1) GNN suffer from the over-smoothing problem and are unable to aggregate deep collaborative signals and (2) GCL adopts a single aggregation paradigm, resulting in a lack of diversity in the feature representation of collaborative signals. To solve the above problems, a multi-view deep graph contrastive learning for collaborative filtering (MD-GCCF) has been proposed from two perspectives. First, a deep graph collaborative signal aggregation module is proposed to learn potential intention similarity representations for deep collaborative signal propagation within a few layers. Second, a novel multi-view contrastive learning module has been proposed, utilizing both local and global contrastive learning views from the collaborative signal aggregation module to enhance deep structures and semantic features in collaborative signals. MD-GCCF improves by 9.52%, 3.34%, and 2.49% compared to the rival models, respectively, in the Amazon book, Yelp2018, and Gowalla datasets.

The implication of sleep disturbances on daily executive functioning and learning problems in children with autism without intellectual disability

BackgroundIndividuals with autism spectrum disorder (ASD) often have sleep disturbances, executive functioning difficulties, and learning problems. The aim of this study was twofold. First, to examine the quality of sleep in children with ASD without intellectual disability and typically developing (TD) children. Second, to explore the implication of sleep disturbances in the association between daily executive functioning and learning problems in children with ASD without intellectual disability. MethodThe sample included 47 children with autism without intellectual disability and 32 TD children. Parent-reported questionnaires were used to assess participants' sleep disturbances, daily executive functions, and learning problems. All statistical analyses performed were adjusted for age, IQ, and medication. ResultsAs expected, the prevalence of sleep disturbances was significantly higher in the ASD group than in the TD group. Regarding the second aim of the study, the total score of sleep disturbances was significantly associated with learning problems and the metacognition subdomain of executive functioning in children with autism without intellectual disability. In turn, executive functions and learning problems were also significantly associated. Finally, sleep disturbances were found to mediate the relationship between global executive functioning and learning problems in children with ASD without intellectual disability. ConclusionsThe findings suggest that sleep disturbances are common in children with autism, and that they play a mediating role in the association between executive functions and learning problems. Therefore, sleep should be a crucial aspect of interventions to improve the learning and academic performance of school-aged children with ASD without intellectual disability.

Asynchronous Federated Learning with Grey Wolf Optimization for the Heterogeneity IoT Devices

The Internet of Things (IoT) creates new options for real-time data collection and Machine Learning (ML) model development. Nonetheless, it's feasible that a particular IoT device does not have sufficient computational power to train and implement a complete learning model. However, there are significant communication costs and data security and privacy concerns associated with sending actual data to a centralised server with a lot of computational power. Federated Learning (FL) is a potential way to train ML models using low-powered devices and Edge Servers (ES) since it is a distributed ML architecture. However, the vast majority of the works in existence make the unsustainable assumption of a synchronized parameters update manner with similar IoT nodes and reliable communications networks. To increase training efficiency and accelerate the speed for heterogeneous IoT devices in an unreliable network environment, we designed an Asynchronous Federated Learning strategy with Grey Wolf Optimization (AFL-GWO) in this research. In particular, we develop a Lightweight Node Selection (LNS) technique and propose an AFL-GWO model to efficiently complete learning tasks. To ensure that diverse IoT nodes with varying computational capabilities and network connectivity are represented in the global learning aggregate, the proposed technique makes node selections on an iterative basis. We show through extensive experiments that our suggested AFL-GWO system outperforms the state-of-the-art techniques on identically and independently distributed (IID) and non-IID data distribution in a variety of contexts.

Masked autoencoders with generalizable self-distillation for skin lesion segmentation.

In the field of skin lesion image segmentation, accurate identification and partitioning of diseased regions is of vital importance for in-depth analysis of skin cancer. Self-supervised learning, i.e., MAE, has emerged as a potent force in the medical imaging domain, which autonomously learns and extracts latent features from unlabeled data, thereby yielding pre-trained models that greatly assist downstream tasks. To encourage pre-trained models to more comprehensively learn the global structural and local detail information inherent in dermoscopy images, we introduce a Teacher-Student architecture, named TEDMAE, by incorporating a self-distillation mechanism, it learns holistic image feature information to improve the generalizable global knowledge learning of the student MAE model. To make the image features learned by the model suitable for unknown test images, two optimization strategies are, Exterior Conversion Augmentation (EC) utilizes random convolutional kernels and linear interpolation to effectively transform the input image into one with the same shape but altered intensities and textures, while Dynamic Feature Generation (DF) employs a nonlinear attention mechanism for feature merging, enhancing the expressive power of the features, are proposed to enhance the generalizability of global features learned by the teacher model, thereby improving the overall generalization capability of the pre-trained models. Experimental results from the three public skin disease datasets, ISIC2019, ISIC2017, and PH indicate that our proposed TEDMAE method outperforms several similar approaches. Specifically, TEDMAE demonstrated optimal segmentation and generalization performance on the ISIC2017 and PH datasets, with Dice scores reaching 82.1% and 91.2%, respectively. The best Jaccard values were 72.6% and 84.5%, while the optimal HD95% values were 13.0% and 8.9%, respectively.

Game on for learning: a holistic exploration of Gamification’s impact on student engagement and academic performance in educational environments

PurposeThe research paper examines the complex relationship between gamification, student engagement and academic performance in educational environments. The study employed a structural equation model that highlights important connections among key constructs within the educational setting.Design/methodology/approachThis research aims to explore the connection between gamification, student engagement and academic performance in educational settings. The study employs various statistical techniques such as factor analysis, Kaiser–Meyer–Olkin (KMO), Bartlett’s test, component transformation matrix, correlation and regression analysis, descriptive statistics, ANOVA, coefficients and coefficient correlations, residual statistics and confirmatory factor analysis (CFA) to analyze the data.FindingsIt was found that active participation by the instructor and good time management skills have a positive impact on student engagement levels (β = 0.380, p < 0.001; β = 0.433 and p < 0.001). However, peer interaction does not significantly predict student engagement (β = −0.068 and p = 0.352). Additionally, there is a positive correlation between student engagement and performance (β = 0.280 and p < 0.001).Research limitations/implicationsThe study highlights the importance of innovative design to fully utilize gamification. Future research should consider design, user characteristics and educational context. The findings can guide informed decisions about gamification in education, fostering motivation and learning objectives.Practical implicationsThe study presents a reliable tool for assessing student engagement and performance in educational settings, demonstrating high Cronbach’s alpha and robust reliability. It identifies student engagement and time management as significant predictors of Global Learning Outcome. The findings can inform decisions on implementing gamification in educational settings, promoting intrinsic motivation and aligning with learning objectives.Social implicationsThe research highlights the transformative impact of gamification on educational practices, highlighting its potential to enhance student experiences, motivate, promote diversity and improve long-term academic performance, highlighting the trend of integrating technology into education.Originality/valueIn today’s ever-changing education landscape, it is essential to incorporate innovative techniques to keep students engaged and enthusiastic about learning. Gamification is one such approach that has become increasingly popular. It is a concept that takes inspiration from the immersive world of games to enhance the overall learning experience.

Global machine learning potentials for molecular crystals.

Molecular crystals are difficult to model with accurate first-principles methods due to large unit cells. On the other hand, accurate modeling is required as polymorphs often differ by only 1kJ/mol. Machine learning interatomic potentials promise to provide accuracy of the baseline first-principles methods with a cost lower by orders of magnitude. Using the existing databases of the density functional theory calculations for molecular crystals and molecules, we train global machine learning interatomic potentials, usable for any molecular crystal. We test the performance of the potentials on experimental benchmarks and show that they perform better than classical force fields and, in some cases, are comparable to the density functional theory calculations.

Context-based local-global fusion network for 3D point cloud classification and segmentation

3D point clouds have gained much research attention because of their ability to represent the spatial information of real-world environments in a detailed manner. Despite recent progress in point cloud processing with deep neural networks, most of them either implement sophisticated local feature aggregation methods or imitate 2D convolution operations in the range of K nearest neighbors with limited local context information. These methods may struggle to distinguish between similar geometric shapes within the local region of K nearest neighbors, such as doors and walls. To address this issue, we propose a novel local–global fusion network that captures the diverse local geometric shapes with global structure information. The proposed local–global fusion network comprises two main modules. Firstly, we have developed an effective approach for local context learning using incremental dilated KNN (IDKNN) as the neighbor selecting mechanism to enlarge the receptive field and incorporate more reliable points for local geometric shape learning. Secondly, a three-direction region-wise spatial attention (TRSA) algorithm has been developed to explore the global contextual dependencies. For global context learning, we first split the entire 3D space into regions with equal numbers of points, and, then, intra-region context features are extracted to learn the inter-region relations from three orthogonal directions, taking global structural knowledge into account. By fusing the local context information and global contextual dependencies, we establish a Local-Global Fusion Network, end-to-end framework, called LGFNet. Extensive experimental results on several benchmark datasets clearly demonstrate our approach can achieve state-of-the-art (SOTA) performance on point cloud classification, part segmentation, and indoor semantic segmentation. In addition, TRSA and IKDNN can be easily used in a plug-and-play fashion with various existing SOTA networks to substantially improve their performance. Our code is available at https://github.com/jasonwjw/IDKNN

Student-Faculty Partnerships as Intercultural Encounters: Co-constructing the Pathways to Global Learning

Student-Faculty Partnerships as Intercultural Encounters: Co-constructing the Pathways to Global Learning

LeGFusion: Locally Enhanced Global Learning for Multimodal Image Fusion

LeGFusion: Locally Enhanced Global Learning for Multimodal Image Fusion

Resformer: Local Frame-Level Feature and Global Segment-Level Feature Joint Learning for Speaker Verification

Resformer: Local Frame-Level Feature and Global Segment-Level Feature Joint Learning for Speaker Verification

Empowering precise advertising with Fed-GANCC: A novel federated learning approach leveraging Generative Adversarial Networks and group clustering.

In the realm of targeted advertising, the demand for precision is paramount, and the traditional centralized machine learning paradigm fails to address this necessity effectively. Two critical challenges persist in the current advertising ecosystem: the data privacy concerns leading to isolated data islands and the complexity in handling non-Independent and Identically Distributed (non-IID) data and concept drift due to the specificity and diversity in user behavior data. Current federated learning frameworks struggle to overcome these hurdles satisfactorily. This paper introduces Fed-GANCC, an innovative federated learning framework that synergizes Generative Adversarial Networks (GANs) and Group Clustering. The framework incorporates a user data augmentation algorithm predicated on adversarial generative networks to enrich user behavior data, curtail the impact of non-uniform data distribution, and enhance the applicability of the global machine learning model. Unlike traditional approaches, our framework offers user data augmentation algorithms based on adversarial generative networks, which not only enriches user behavior data but also reduces the challenges posed by non-uniform data distribution, thereby enhancing the applicability of the global machine learning (ML) model. The effectiveness of Fed-GANCC is distinctly showcased through experimental results, outperforming contemporary methods like FED-AVG and FED-SGD in terms of accuracy, loss value, and receiver operating characteristic (ROC) indicators within the same computing time. Experimental results vindicate the effectiveness of Fed-GANCC, revealing substantial enhancements in accuracy, loss value, and receiver operating characteristic (ROC) metrics compared to FED-AVG and FED-SGD given the same computational time. These outcomes underline Fed-GANCC's exceptional prowess in mitigating issues such as isolated data islands, non-IID data, and concept drift. With its novel approach to addressing the prevailing challenges in targeted advertising such as isolated data islands, non-IID data, and concept drift, the Fed-GANCC framework stands as a benchmark, paving the way for future advancements in federated learning solutions tailored for the advertising domain. The Fed-GANCC framework promises to offer pivotal insights for the future development of efficient and advanced federated learning solutions for targeted advertising.

Ideals versus realities of learning poverty and human rights

PurposeThis paper will identify and further explore the ideals versus realities of learning poverty and the consequential effects on our moral obligations and responsibilities. The wealthy nations are now under further pressure to recognise and realise their moral obligations to enabling social justice in the context of access, and distribution, of vaccines for the poorer nations. Learning poverty has always been a feature of our global economic, and institutional order, and has become an increasingly important factor in achieving justice.Design/methodology/approachThe study focusses on a human rights approach to learning poverty and the ideals versus the realities of what we are beginning to see in the times of a global pandemic. The major challenges to justice is inherent to the recognition that wealthy nations continue to have a pivotal role in the reduction of poverty. The identified major challenges in the context of learning poverty are: “nation states and the global pandemic”, “international interactions and learning poverty” and “global institutions and learning inequalities”. In particular, the authors explore the concept of ideals versus realities through three “challenges”, which continues to challenge any semblance of justice in the current global vaccine distribution. Nation states and borders, international interactions and global institutions remain barriers in overcoming what is becoming a reality of learning poverty.FindingsThis paper seeks to look beyond the economics of vaccine trade and seek a way to accept a moral claim of justice for all. The authors consider how wealthy nations are active participants in the emergence of learning poverty for many nations.Originality/valueBy exploring the ideals versus realities of learning poverty, and human rights, the authors highlight some of the challenges, and wealthy nations moral obligations, through the emergence of a new dimensional indicator of poverty, learning poverty.

A Pragmatic Approach to the Prevention of Preterm Birth

Preterm birth (PTB) -delivery before 37 weeks of gestation- occurs in ~8% of Canadian pregnancies, a rate that has remained stable for >20 years. Prematurity remains the leading cause of neonatal mortality- especially at the extreme of viability: 22–26 weeks gestation. Children born prematurely have increased rates of cerebral palsy, global neurodevelopmental impairment, learning disabilities, deafness, vision impairment and behavioural diagnoses. Hospital length of stay and re-admission rates in the first year of life are higher in babies born preterm. The family impact is immeasurable in the domains of mental health diagnoses, financial health, and disordered social interactions. Prematurity has a longitudinal impact on reproductive and public health with an increased risk of occurrence in subsequent pregnancies.

Enhancing Local–Global Representation Learning for Image Restoration

Enhancing Local–Global Representation Learning for Image Restoration

LGABL: UHD Multi-Exposure Image Fusion via Local and Global Aware Bilateral Learning

LGABL: UHD Multi-Exposure Image Fusion via Local and Global Aware Bilateral Learning

Semantic segmentation using cross-stage feature reweighting and efficient self-attention

Recently, vision transformers have demonstrated strong performance in various computer vision tasks. The success of ViTs can be attribute to the ability of capturing long-range dependencies. However, transformer-based approaches often yield segmentation maps with incomplete object structures because of restricted cross-stage information propagation and lack of low-level details. To address these problems, we introduce a CNN-transformer semantic segmentation architecture which adopts a CNN backbone for multi-level feature extraction and a transformer encoder that focuses on global perception learning. Transformer embeddings of all stages are integrated to compute feature weights for dynamic cross-stage feature reweighting. As a result, high-level semantic context and low-level spatial details can be embedded into each stage to preserve multi-level information. An efficient attention-based feature fusion mechanism is developed to combine reweighted transformer embeddings with CNN features to generate segmentation maps with more complete object structure. Different from regular self-attention that has quadratic computational complexity, our efficient self-attention method achieves similar performance with linear complexity. Experimental results on ADE20K and Cityscapes datasets show that the proposed segmentation approach demonstrates superior performance against most state-of-the-art networks.

Deep local-to-global feature learning for medical image super-resolution

Medical images play a vital role in medical analysis by providing crucial information about patients’ pathological conditions. However, the quality of these images can be compromised by many factors, such as limited resolution of the instruments, artifacts caused by movements, and the complexity of the scanned areas. As a result, low-resolution (LR) images cannot provide sufficient information for diagnosis. To address this issue, researchers have attempted to apply image super-resolution (SR) techniques to restore the high-resolution (HR) images from their LR counterparts. However, these techniques are designed for generic images, and thus suffer from many challenges unique to medical images. An obvious one is the diversity of the scanned objects; for example, the organs, tissues, and vessels typically appear in different sizes and shapes, and are thus hard to restore with standard convolution neural networks (CNNs). In this paper, we develop a dynamic-local learning framework to capture the details of these diverse areas, consisting of deformable convolutions with adjustable kernel shapes. Moreover, the global information between the tissues and organs is vital for medical diagnosis. To preserve global information, we propose pixel–pixel and patch–patch global learning using a non-local mechanism and a vision transformer (ViT), respectively. The result is a novel CNN-ViT neural network with Local-to-Global feature learning for medical image SR, referred to as LGSR, which can accurately restore both local details and global information. We evaluate our method on six public datasets and one large-scale private dataset, which include five different types of medical images (i.e., Ultrasound, OCT, Endoscope, CT, and MRI images). Experiments show that the proposed method achieves superior PSNR/SSIM and visual performance than the state of the arts with competitive computational costs, measured in network parameters, runtime, and FLOPs. What is more, the experiment conducted on OCT image segmentation for the downstream task demonstrates a significantly positive performance effect of LGSR.

AdaGCR: An improved method for optimizing machine learning training process

In contemporary machine learning, training datasets are typically divided into batches, and models are updated incrementally through batch iterations to save memory and reduce overfitting. However, determining the optimal hyperparameters like learning rate, batch size and number of epochs remains a challenge which often relying on empirical insights. This paper explores a novel method called Adaptive Gradient Conflict Rate (AdaGCR) to optimize the training process. It leverages the idea of gradient conflict rate, which reflects the models position within a batch model set and accordingly adjusts the global learning rate. This proposed method is tested by training a Deep Neural Network (DNN) model with MNIST dataset which represents simple tasks and a ResNet-18 model with CIFAR-10 dataset which represents more complicated tasks for solving real world problems. Experiments conducted on DNN demonstrates the proposed methods effectiveness in reducing overfitting and enhancing convergence, particularly with a well-suited initial learning rate. However, its applicability to more complex models like ResNet-18 may require further refinements, such as layer-specific learning rate adjustments. Future research should focus on fine-tuning AdaGCR and extending its utility across diverse machine learning models and tasks.