Learning Training Research Articles

The effects of language learning on cognitive functioning and psychosocial well-being in cognitively healthy older adults: A semi-blind randomized controlled trial

ABSTRACT This study investigated the impact of language learning in comparison to other complex learning activities on cognitive functioning and psychosocial well-being in cognitively healthy, community-dwelling older adults. In a randomized controlled trial, 43 Dutch functionally monolinguals aged 65–78 completed a three-month English course (n = 15), music training (n = 13), or a lecture series (n = 15). Cognitive functioning (global cognition, cognitive flexibility, episodic memory, working memory, verbal fluency, and attention) and psychosocial well-being were assessed before and immediately after the intervention, and at a four-month follow-up. The language learners significantly improved on episodic memory and cognitive flexibility. However, the magnitude of cognitive change did not significantly differ between the language learning and music training conditions, except for a larger positive change in cognitive flexibility for the language learners from pretest to follow-up. Our results suggest that language learning in later life can improve some cognitive functions and fluency in the additional language, but that its unique effects seem limited.

Hard sample mining-enabled supervised contrastive feature learning for wind turbine pitch system fault diagnosis

Abstract The presence of multiple failure severities in the wind turbine pitch system due to the long-term wear and tear poses challenges in accurately classifying the system’s health condition, thus increasing maintenance costs or damage risks. This paper proposes a novel method based on hard sample mining (HSM)-enabled supervised contrastive learning to address this problem. The proposed method leverages the powerful feature extraction capabilities of supervised contrastive learning to extract discriminative features from highly imbalanced data. Additionally, the method incorporates a cosine similarity-based HSM framework that constructs hard sample pairs within mini-batches during both the representation learning and classifier training phases, thereby improving the diagnostic performance of the model for hard samples. The proposed method achieves macro G-mean of 0.9991 and 0.9971 on two real datasets containing data on wind turbine pitch system cog belt fractures. These results indicate significantly superior fault diagnosis performance compared to existing methods, highlighting its potential for enhancing the reliability and efficiency of fault diagnosis in wind turbine pitch systems.

Numerical simulation of wormhole propagation in fractured carbonate rocks during acidizing using a simplified Stokes–Brinkman model

Most numerical simulations for modeling acid reactive fluid transport and wormhole propagation during matrix acidizing, waterflooding, and CO2 sequestration in carbonate formations are computationally expensive, limiting real-time reservoir management and deep learning training datasets generation for inverse modeling research. Therefore, there is a need for less computationally expensive acid-reactive fluid flow models with adequate accuracy. This study developed and validated a simplified acid reactive-transport model by integrating a simplified Stokes–Brinkman model (as opposed to Darcy’s law), an averaged continuum model, and a pseudo-fracture model. Using FEniCS, the model effectively simulates acid-reactive fluid transport and wormhole propagation in carbonate rocks, achieving a high R-square value of about 0.97 based on a quantitative comparison of the breakthrough volume with other models. The simplified model can also simulate wormhole propagation for the reciprocal of the Damköhler number (1/Da) ranging from 0.001 to 1 with adequate accuracy. Sensitivity studies on the natural fracture parameters such as orientation, length, width, and density showed that higher fracture density, wider fracture aperture, longer fracture length, and orientation aligned with the direction of acid injection contribute to lower pore volume to breakthrough ratio but may not increase long-term acid stimulation efficiency. Also, the presence or absence of fractures in the matrix does not alter the dissolving patterns and optimum injection rate. This simple acid reactive-transport model can generate large training datasets for developing surrogate models in deep learning research. Finally, the FEniCS code in this paper is shared so future researchers can reproduce the results or extend the research work.

Formability Prediction Using Machine Learning Combined with Process Design for High-Drawing-Ratio Aluminum Alloy Cups.

Deep drawing has been practiced in various manufacturing industries for many years. With the aid of stamping equipment, materials are sheared to different shapes and dimensions for users. Meanwhile, through artificial intelligence (AI) training, machines can make decisions or perform various functions. The aim of this study is to discuss the geometric and process parameters for A7075 in deep drawing and derive the formable regions of sound products for different forming parameters. Four parameters-forming temperature, punch speed, blank diameter and thickness-are used to investigate their effects on the forming results. Through finite element simulation, a database is established and used for machine learning (ML) training and validation to derive an AI prediction model. Importing the forming parameters into this prediction model can obtain the forming results rapidly. To validate the formable regions of sound products, several experiments are conducted and the results are compared with the prediction results to verify the feasibility of applying ML to deep drawing processes of aluminum alloy A7075 and the reliability of the AI prediction model.

Bird song comparison using deep learning trained from avian perceptual judgments.

Our understanding of bird song, a model system for animal communication and the neurobiology of learning, depends critically on making reliable, validated comparisons between the complex multidimensional syllables that are used in songs. However, most assessments of song similarity are based on human inspection of spectrograms, or computational methods developed from human intuitions. Using a novel automated operant conditioning system, we collected a large corpus of zebra finches' (Taeniopygia guttata) decisions about song syllable similarity. We use this dataset to compare and externally validate similarity algorithms in widely-used publicly available software (Raven, Sound Analysis Pro, Luscinia). Although these methods all perform better than chance, they do not closely emulate the avian assessments. We then introduce a novel deep learning method that can produce perceptual similarity judgements trained on such avian decisions. We find that this new method outperforms the established methods in accuracy and more closely approaches the avian assessments. Inconsistent (hence ambiguous) decisions are a common occurrence in animal behavioural data; we show that a modification of the deep learning training that accommodates these leads to the strongest performance. We argue this approach is the best way to validate methods to compare song similarity, that our dataset can be used to validate novel methods, and that the general approach can easily be extended to other species.

Stochastic minibatch approach to the ptychographic iterative engine

The ptychographic iterative engine (PIE) is a widely used algorithm that enables phase retrieval at nanometer-scale resolution over a wide range of imaging experiment configurations. By analyzing diffraction intensities from multiple scanning locations where a probing wavefield interacts with a sample, the algorithm solves a difficult optimization problem with constraints derived from the experimental geometry as well as sample properties. The effectiveness at which this optimization problem is solved is highly dependent on the ordering in which we use the measured diffraction intensities in the algorithm, and random ordering is widely used due to the limited ability to escape from stagnation in poor-quality local solutions. In this study, we introduce an extension to the PIE algorithm that uses ideas popularized in recent machine learning training methods, in this case minibatch stochastic gradient descent. Our results demonstrate that these new techniques significantly improve the convergence properties of the PIE numerical optimization problem.

Experts and Expertise in Organizations: An Integrative Review on Individual Expertise

Experts and expertise contribute to consequential organizational decisions from recruitment to CEO succession, but these constructs are inconsistently operationalized and poorly understood. To better explicate how experts and expertise function in organizations, we first conduct an integrative review of the general literature to describe what is known about these phenomena in cognitive science, psychology, and the clinical and technical professions. This review of the general literature indicates that expertise represents domain-specific hierarchical knowledge structures developed by an individual over time. The quality of the individual's domain-related education, training, and opportunities for practice and learning affect the level of expertise acquired. We then review what is known about experts and expertise in organizations. Many organizational studies on expertise focus on an individual's years of experience rather than the nature of that experience or its contribution to expertise. Conflating expertise with years of experience generally leads to less consistent effects on performance than operationalizing expertise in terms of individual cognitive processes, knowledge, and capabilities. Findings from organizational studies that do assess expertise are in line with the general literature, indicating that the quality of practice and learning experiences are particularly important to developing expertise. We then offer ways for scholars to better study how expertise functions in organizations and conclude by developing implications for practice.

BIDSAlign: a library for automatic merging and preprocessing of multiple EEG repositories

Objective.This study aims to address the challenges associated with data-driven electroencephalography (EEG) data analysis by introducing a standardised library calledBIDSAlign. This library efficiently processes and merges heterogeneous EEG datasets from different sources into a common standard template. The goal of this work is to create an environment that allows to preprocess public datasets in order to provide data for the effective training of deep learning (DL) architectures.Approach.The library can handle both Brain Imaging Data Structure (BIDS) and non-BIDS datasets, allowing the user to easily preprocess multiple public datasets. It unifies the EEG recordings acquired with different settings by defining a common pipeline and a specified channel template. An array of visualisation functions is provided inside the library, together with a user-friendly graphical user interface to assist non-expert users throughout the workflow.Main results.BIDSAlign enables the effective use of public EEG datasets, providing valuable medical insights, even for non-experts in the field. Results from applying the library to datasets from OpenNeuro demonstrate its ability to extract significant medical knowledge through an end-to-end workflow, facilitating group analysis, visual comparison and statistical testing.Significance.BIDSAlign solves the lack of large EEG datasets by aligning multiple datasets to a standard template. This unlocks the potential of public EEG data for training DL models. It paves the way to promising contributions based on DL to clinical and non-clinical EEG research, offering insights that can inform neurological disease diagnosis and treatment strategies.

Entropy‐based sampling for efficient training of deep learning on CNC machining dataset

AbstractIn the domain of modern manufacturing, computer numerical control (CNC) milling machines have emerged as instrumental assets. However, the data they generate is of vast amount, but usually contains redundancies and displays consistent patterns, making it inefficient for deep learning training. This paper proposes a novel sampling algorithm tailored for CNC milling machine data, emphasizing both diversity and efficiency. The proposed method leverages the entropy concept from the information‐theoretic perspective to evaluate and enhance data diversity, aiming to achieve efficient learning with high accuracy. This in turn enables to not only facilitates a deeper understanding of CNC data characteristics but also contributes significantly to the optimization of deep learning training processes in the context of CNC milling data.

Transforming Teaching Practices Through Comprehensive Clinical Supervision in Secondary Schools

The teachers do not agree that the principal's supervision is more bureaucratic and autocratic, emphasizing hierarchy. The fact is that 44% of teachers do not like the supervision carried out in the classroom by either the principal or his deputy, and 58.8% of secondary school teachers disagree with teaching observations. This is urgent to research because there is a difference between theory and reality. This research aimed to evaluate the effect of clinical supervision on the teaching performance of teachers in secondary schools. Quantitative research method using a combination of teaching and learning observation instruments that have been adapted from all schools in assessing. The subjects were 38 secondary school teachers. Data collection techniques using observation instruments. The instrument was developed from indicators used as a tool to measure teacher performance. The instrument is given a score from point 1 to point 5. Analysis technique using SPSS Version 25.0 with measurements based on Learning Plans, induction tools, material delivery, questioning techniques, student involvement, reinforcement, training, and student assignments, checking training and student assignments, closing learning, and classroom management before and after clinical supervision. The results and findings of this research help teachers in schools to know the strengths and weaknesses when teaching is delivered in class. Clinical supervision also helps teachers improve the teaching and learning process more effectively so that teaching can improve student mastery. Clinical supervision can also be used as a guide for high school students to improve their knowledge.

Mapping educators’ insights of student problem-solving difficulties while implementing video-assisted self-regulated learning training in classrooms

Abstract Developing self-regulated learning (SRL) skills among young adolescents is crucial for lifelong learning, and teachers play a vital role in fostering these skills. Problem-solving is a key SRL skill; however, both learners and teachers encounter challenges during the problem-solving process and in evaluating its incorporated strategies. To tackle these challenges, this study examined students’ problem-solving difficulties using a generic video-assisted SRL training programme, based on teachers’ reports, and analysed these challenges across STEM and non-STEM domains. This study focusses on identifying and classifying the difficulties of 241 students within the problem-solving framework, as reported by teachers across various disciplines and grades (6–11). The analysis revealed that the most significant difficulties arose during the problem identification, followed by time planning and management, problem exploration, solution development and solution evaluation. Specifically, STEM teachers reported relatively fewer difficulties in problem exploration than in other steps, while non-STEM teachers reported fewer challenges in presenting and communicating solutions to peers. A dedicated map named the problem-solving sort out setbacks (PS-SOS) map was created to pinpoint challenges within the problem-solving process. The application of this mapping technique can further support the development of technology-based feedback systems, including digital assistants, which offer valuable assistance to students.

Exploring the Frontiers of Synthetic Image-Based Deep Learning Training in Digital X-ray Radiography

Exploring the Frontiers of Synthetic Image-Based Deep Learning Training in Digital X-ray Radiography

Integrating international Chinese visualization teaching and vocational skills training: leveraging attention-connectionist temporal classification models.

The teaching of Chinese as a second language has become increasingly crucial for promoting cross-cultural exchange and mutual learning worldwide. However, traditional approaches to international Chinese language teaching have limitations that hinder their effectiveness, such as outdated teaching materials, lack of qualified instructors, and limited access to learning facilities. To overcome these challenges, it is imperative to develop intelligent and visually engaging methods for teaching international Chinese language learners. In this article, we propose leveraging speech recognition technology within artificial intelligence to create an oral assistance platform that provides visualized pinyin-formatted feedback to learners. Additionally, this system can identify accent errors and provide vocational skills training to improve learners' communication abilities. To achieve this, we propose the Attention-Connectionist Temporal Classification (CTC) model, which utilizes a specific temporal convolutional neural network to capture the location information necessary for accurate speech recognition. Our experimental results demonstrate that this model outperforms similar approaches, with significant reductions in error rates for both validation and test sets, compared with the original Attention model, Claim, Evidence, Reasoning (CER) is reduced by 0.67%. Overall, our proposed approach has significant potential for enhancing the efficiency and effectiveness of vocational skills training for international Chinese language learners.

Mastering English Together: Twinkl Educational Publishing Training for Teachers and Young Learners at KB & TK Al - Madani Malang

Mastering English Together: Twinkl Educational Publishing Training for Teachers and Young Learners at KB & TK Al - Madani Malang

Construction of Graduate Talent Training Mechanism based on Multi-Disciplinary Crossing and Integration

It is an urgent requirement for the high-quality development of universities and society to build an effective graduate talent training mechanism. The importance of building a graduate talent training mechanism based on multi-disciplinary crossing and integration is discussed. Under the background of "double first-class" construction, innovation ability has become the key to measure the quality of graduate training, and multi-disciplinary cross-training is an important way to improve this ability. To optimize the environment of interdisciplinary integration, the author puts forward some strategies, such as the integration of science and education, the cross-disciplinary training mode, the improvement of interdisciplinary course elective mechanism, the construction of the training ideological mechanism of deep integration of industry, university and research, the establishment of teaching evaluation system and the creation of the training mode of "reality teaching and learning training". At the same time, it is necessary to strengthen the construction of school-enterprise collaborative graduate training mode, including resolving the conflict of values and interest pursuit, strengthening the top-level design, and co-construction evaluation mechanism. It has a certain reference significance for the mechanism of graduate personnel training.

Automatic rating of incomplete hippocampal inversions evaluated across multiple cohorts

Incomplete Hippocampal Inversion (IHI), sometimes called hippocampal malrotation, is an atypical anatomical pattern of the hippocampus found in about 20% of the general population. IHI can be visually assessed on coronal slices of T1 weighted MR images, using a composite score that combines four anatomical criteria. IHI has been associated with several brain disorders (epilepsy, schizophrenia). However, these studies were based on small samples. Furthermore, the factors (genetic or environmental) that contribute to the genesis of IHI are largely unknown. Large-scale studies are thus needed to further understand IHI and their potential relationships to neurological and psychiatric disorders. However, visual evaluation is long and tedious, justifying the need for an automatic method. In this paper, we propose, for the first time, to automatically rate IHI. We proceed by predicting four anatomical criteria, which are then summed up to form the IHI score, providing the advantage of an interpretable score. We provided an extensive experimental investigation of different machine learning methods and training strategies. We performed automatic rating using a variety of deep learning models (”conv5-FC3”, ResNet and ”SECNN”) as well as a ridge regression. We studied the generalization of our models using different cohorts and performed multi-cohort learning. We relied on a large population of 2,008 participants from the IMAGEN study, 993 and 403 participants from the QTIM and QTAB studies as well as 985 subjects from the UKBiobank. We showed that deep learning models outperformed a ridge regression. We demonstrated that the performances of the ”conv5-FC3” network were at least as good as more complex networks while maintaining a low complexity and computation time. We showed that training on a single cohort may lack in variability while training on several cohorts improves generalization (acceptable performances on all tested cohorts including some that are not included in training). The trained models will be made publicly available should the manuscript be accepted.

Learners use of audio/video playback controls in technology-enhanced listening: a systematic review

This systematic review investigates learner use of audio/video playback (AVP) controls in technology-enhanced listening environments. To this aim, 61 academic works produced from 2000–2021 underwent inclusion/exclusion criteria and were analyzed. The resulting corpus was made up of 16 peer-reviewed articles. We first situate the studies examined with regards to contextual, educational, and methodological characteristics of AVP controls research. Then, we conducted thematic analysis to identify, analyze and report affordances and limitations of AVP controls. Four affordances and two limitations were identified. The affordances describe how AVP controls enable students to resort to different strategies, change their interaction patterns with learning materials, address comprehension problems and enhance listening comprehension performance. Similarly, limitations describe how proficiency level restricts interaction with AVP controls and the need for learner training on AVP buttons. Results are discussed along with integrated data. Pedagogical implications and avenues for further research are also discussed.

Research on deep learning restoration algorithm of X-ray backscatter imaging based on virtual training dataset

The X-ray lobster eye lens, an innovative technique for focusing high-energy radiation, enables wide-field X-ray imaging. However, its inherent cross point spread function introduces noise and degradation into the resultant images. Conventional image restoration methods are inadequate for suppressing such noise. This paper introduces a backscatter image restoration technique utilizing a virtual training dataset. By convolving the point spread function (PSF) with an object to simulate the image degradation process, the method generates a multitude of convolved images for deep learning training, eliminating the need for manual annotation. Given the high structural similarity between the synthetic convolved images and actual backscatter images, the trained model effectively restores real backscatter images. The restoration process yields a structural similarity index (SSIM) of 0.86 and a mean intersection over union (MIoU) of 0.83 when compared to the reference images. This approach mitigates the limitations of sparse real backscatter datasets, substantially reducing image acquisition time, decreasing radiation flux, and enhancing system safety.

Design and Application of an Onboard Particle Identification Platform Based on Convolutional Neural Networks

Space radiation particle detection plays a crucial role in scientific research and engineering practice, especially in particle species identification. Currently, commonly used in-orbit particle identification techniques include telescope methods, electrostatic analysis time of flight (ESA × TOF), time-of-flight energy (TOF × E), and pulse shape discrimination (PSD). However, these methods usually fail to utilize the full waveform information containing rich features, and their particle identification results may be affected by the random rise and fall of particle deposition and noise interference. In this study, a low-latency and lightweight onboard FPGA real-time particle identification platform based on full waveform information was developed utilizing the superior target classification, robustness, and generalization capabilities of convolutional neural networks (CNNs). The platform constructs diversified input datasets based on the physical features of waveforms and uses Optuna and Pytorch software architectures for model training. The hardware platform is responsible for the real-time inference of waveform data and the dynamic expansion of the dataset. The platform was utilized for deep learning training and the testing of the historical waveform data of neutron and gamma rays, and the inference time of a single waveform takes 4.9 microseconds, with an accuracy rate of over 97%. The classification expectation FOM (figure-of-merit) value of this CNN model is 133, which is better than the traditional pulse shape discrimination (PSD) algorithm’s FOM value of 0.8. The development of this platform not only improves the accuracy and efficiency of space particle discrimination but also provides an advanced tool for future space environment monitoring, which is of great value for engineering applications.

Symmetric silicon microring resonator optical crossbar array for accelerated inference and training in deep learning

Photonic integrated circuits are emerging as a promising platform for accelerating matrix multiplications in deep learning, leveraging the inherent parallel nature of light. Although various schemes have been proposed and demonstrated to realize such photonic matrix accelerators, the in situ training of artificial neural networks using photonic accelerators remains challenging due to the difficulty of direct on-chip backpropagation on a photonic chip. In this work, we propose a silicon microring resonator (MRR) optical crossbar array with a symmetric structure that allows for simple on-chip backpropagation, potentially enabling the acceleration of both the inference and training phases of deep learning. We demonstrate a 4×4 circuit on a Si-on-insulator platform and use it to perform inference tasks of a simple neural network for classifying iris flowers, achieving a classification accuracy of 93.3%. Subsequently, we train the neural network using simulated on-chip backpropagation and achieve an accuracy of 91.1% in the same inference task after training. Furthermore, we simulate a convolutional neural network for handwritten digit recognition, using a 9×9 MRR crossbar array to perform the convolution operations. This work contributes to the realization of compact and energy-efficient photonic accelerators for deep learning.