Learning Criterion Research Articles

Brain areas activated during visual learning in the cichlid fish Pseudotropheus zebra

The neural correlates of most cognitive functions in fish are unknown. This project aimed to identify brain regions involved in visual learning in the cichlid fish Pseudotropheus zebra. The expression of the protein pS6 was measured in 19 brain areas and compared between groups of individuals subjected to four different behavioral contexts (control, avoidance, trained, and novelty groups). Control group individuals were sacrificed with minimal interactions. Fish in the avoidance group were chased with a net for an hour, after which they were sacrificed. Individuals in the trained group received daily training sessions to associate a visual object with a food reward. They were sacrificed the day they reached learning criterion. Fish in the novelty group were habituated to one set of visual stimuli, then faced a change in stimulus type (novelty stimulus) before they were sacrificed. Fish in the three treatment groups showed the largest activation of pS6 in the inferior lobes and the tectum opticum compared to the control group. The avoidance group showed additional activation in the preoptic area, several telencephalic regions, the torus semicircularis, and the reticular formation. The trained group that received a food reward, showed additional activation of the torus lateralis, a tertiary gustatory center. The only area that showed strong activation in all three treatment groups was the nucleus diffusus situated within the inferior lobe. The inferior lobe receives prominent visual input from the tectum via the nucleus glomerulosus but so far, nothing is known about the functional details of this pathway. Our study showed for the first time that the inferior lobes play an important role in visual learning and object recognition.

Cost-Sensitive YOLOv5 for Detecting Surface Defects of Industrial Products.

Owing to the remarkable development of deep learning algorithms, defect detection techniques based on deep neural networks have been extensively applied in industrial production. Most existing surface defect detection models assign equal costs to the classification errors among different defect categories but do not strictly distinguish them. However, various errors can generate a great discrepancy in decision risk or classification costs and then produce a cost-sensitive issue that is crucial to the manufacturing process. To address this engineering challenge, we propose a novel supervised classification cost-sensitive learning method (SCCS) and apply it to improve YOLOv5 as CS-YOLOv5, where the classification loss function of object detection was reconstructed according to a new cost-sensitive learning criterion explained by a label-cost vector selection method. In this way, the classification risk information from a cost matrix is directly introduced into the detection model and fully exploited in training. As a result, the developed approach can make low-risk classification decisions for defect detection. It is applicable for direct cost-sensitive learning based on a cost matrix to implement detection tasks. Using two datasets of a painting surface and a hot-rolled steel strip surface, our CS-YOLOv5 model outperforms the original version with respect to cost under different positive classes, coefficients, and weight ratios, but also maintains effective detection performance measured by mAP and F1 scores.

Enhancing Safety Training Performance Using Extended Reality: A Hybrid Delphi–AHP Multi-Attribute Analysis in a Type-2 Fuzzy Environment

Safety training effectively addresses the inexperience of and lack of knowledge among construction workers, which are some of the most significant contributors to workplace accidents on construction sites. This paper aims to understand the effectiveness of different extended reality (XR) technologies in imparting important construction safety training to construction workers in a virtual environment compared to conventional classroom training sessions. A group of experts were engaged to understand the most effective learning criteria and the impact of XR visualizations, and their responses were analysed using the interval type-2 fuzzy Delphi (IT2FD) method. Following this, a cohort of engineering students were subjected to construction safety training in traditional, augmented reality (AR) and virtual reality (VR) environments. Their feedback was collected using an online questionnaire and the responses were analysed using the interval type-2 fuzzy analytic hierarchy process (IT2F–AHP). The results revealed that addressing the virtual interface design of the training to maintain the attention of trainees and ensuring the virtual environment’s resemblance to the actual site and task were the most important factors in ensuring effective knowledge retention by the trainees. AR visualizations were most effective at imparting knowledge, and their interactive nature allowed trainees to retain the learned knowledge.

Robust state of charge estimation of lithium-ion battery via mixture kernel mean p-power error loss LSTM with heap-based-optimizer

The state of charge (SOC) estimation of lithium-ion battery is an important function in the battery management system (BMS) of electric vehicles. The long short term memory (LSTM) model can be employed for SOC estimation, which is capable of estimating the future changing states of a nonlinear system. Since the BMS usually works under complicated operating conditions, i.e the real measurement data used for model training may be corrupted by non-Gaussian noise, and thus the performance of the original LSTM with the mean square error (MSE) loss may deteriorate. Therefore, a novel LSTM with mixture kernel mean p-power error (MKMPE) loss, called MKMPE-LSTM, is developed by using the MKMPE loss to replace the MSE as the learning criterion in LSTM framework, which can achieve robust SOC estimation under the measurement data contaminated with non-Gaussian noises (or outliers) because of the MKMPE containing the p-order moments of the error distribution. In addition, a meta-heuristic algorithm, called heap-based-optimizer (HBO), is employed to optimize the hyper-parameters (mainly including learning rate, number of hidden layer neuron and value of p in MKMPE) of the proposed MKMPE-LSTM model to further improve its flexibility and generalization performance, and a novel hybrid model (HBO-MKMPE-LSTM) is established for SOC estimation under non-Gaussian noise cases. Finally, several tests are performed under various cases through a benchmark to evaluate the performance of the proposed HBO-MKMPE-LSTM model, and the results demonstrate that the proposed hybrid method can provide a good robustness and accuracy under different non-Gaussian measurement noises, and the SOC estimation results in terms of mean square error (MSE), root MSE(RMSE), mean absolute relative error (MARE), and determination coefficient R2 are less than 0.05%, 3%, 3%,and above 99.8% at 25℃, respectively.

Digital Flash Card Media for Early Reading Learning in Elementary Schools

Many students need to learn the letters correctly, often make mistakes in distinguishing letters that are almost similar in shape, have difficulty, and are not fluent in reading. For this reason, teachers need to develop learning media that can attract students' interest, are practical to use, digital-based, and are by student characteristics and technological developments so that learning to read becomes fun. This study aimed to develop digital Flash Card media in early reading learning in elementary schools. The subjects in this study were first-grade students, totaling 30 students. The research method used is the method of research and development or Research and Development (R&D). Data collection was carried out through interviews, questionnaires, and observation. The collected data were analyzed using qualitative and quantitative descriptive analysis techniques. The results showed that the validation test for digital Flash Card media products by material experts was 87% in a very valid category. The assessment by media experts is 90% with a very valid category, and an evaluation by linguists is 89% with a very valid type. Furthermore, the results of observing trials using digital Flash Card media on a large scale for elementary school students obtained an average of 87.7% in the practical category. It was concluded that the development of digital Flash Card media met the valid, helpful, and effective criteria for learning to read in elementary schools.

Optimal Transport for Unsupervised Denoising Learning.

Recently, much progress has been made in unsupervised denoising learning. However, existing methods more or less rely on some assumptions on the signal and/or degradation model, which limits their practical performance. How to construct an optimal criterion for unsupervised denoising learning without any prior knowledge on the degradation model is still an open question. Toward answering this question, this work proposes a criterion for unsupervised denoising learning based on the optimal transport theory. This criterion has favorable properties, e.g., approximately maximal preservation of the information of the signal, whilst achieving perceptual reconstruction. Furthermore, though a relaxed unconstrained formulation is used in practical implementation, we prove that the relaxed formulation in theory has the same solution as the original constrained formulation. Experiments on synthetic and real-world data, including realistic photographic, microscopy, depth, and raw depth images, demonstrate that the proposed method even compares favorably with supervised methods, e.g., approaching the PSNR of supervised methods while having better perceptual quality. Particularly, for spatially correlated noise and realistic microscopy images, the proposed method not only achieves better perceptual quality but also has higher PSNR than supervised methods. Besides, it shows remarkable superiority in harsh practical conditions with complex noise, e.g., raw depth images. Code is available at https://github.com/wangweiSJTU/OTUR.

Creating a Virtual (Digital) Educational Environment for Teaching Biology in the Conditions of Technoparks and Quantoriums at School and University

The article describes the features of the interaction of a general education school and a technopark of a pedagogical university in the implementation of modern biological education. The relationship of biology with other sciences is considered, the types of educational technologies used in the process of interaction are presented, the significance of the digital laboratories of the technopark are revealed. The article considers as a scientific problem the identification of promising areas of didactic research, the expansion of their research field. In the article, the provisions of general didactics are concretized in relation to new objects included in its research field, the problems of didactics of higher education are considered. Also, the article deals with virtual reality, more precisely, its concept for creating a more powerful and useful virtual learning environment than existing approaches to learning based on traditional technologies, by incorporating new technologies and new learning criteria. A method of vector modeling of virtual reality, presented for use in various types of self-education, based on information and communication technologies, educational technologies for blended teaching of biology in high school is proposed and substantiated.

Research on renewal and transformation of urban negative space based on AI spatial identification technology

This paper aims to study the renewal and transformation of urban negative space based on AI spatial identification technology, and takes Deya Road in Changsha as a typical case for field investigation. On the basis of data analysis based on the establishment of the standard model, through the feature analysis of Deya Road negative space and the data analysis of the standard model, combined with AI image space recognition technology and algorithm-based logical thinking, the paper summarizes the spatial feature points of the negative space for digital translation, providing data sources and evaluation criteria for AI machine learning. At the same time, the probability of success of spatial identification and spatial model library are established. Combining with the record of quantitative data, the positive countermeasures are put forward for the negative spatial remodeling of streets.

Digital competition and capabilities' augmentation on international cohorts: evidence from the 'case competition 2019'

Intertwining the fields of education policy research (EPR) and educational experiment research (EER), this investigation exploits the usage of digital technologies and team-competition to test whether the phenomenon of digital competition yields a divergent (and likely more positive) impact than traditional classroom coursework, measurable on the degree of learning (criterion 1) and satisfaction (criterion 2) within a target population of international higher education students (HES). An event fashioned and simply dubbed as 'case competition', provided the context and virtual environment for a quasi-experiment which assigned participants (the experiment group) to seven teams competing to solve an industry case. Primary data obtained from the latter through an individual anonymised event's feedback-questionnaire was crossed-checked against a dataset of the overall annual students' evaluation of education quality (SEEQ) instrument; the respondents constitute the opposing control group. Results unravelled hidden benefits of endowing classroom-instruction with this off-stream approach. Furthermore, this led to the modelling of the digital competition augmented learning (DCAL) framework merging the virtues of three main bodies of educational theory: student-centric learning (SCL); problem-based learning (PBL) and technology-enhanced learning (TEL).

Image Super-Resolution using Generative Adversarial Networks with EfficientNetV2

The image super-resolution is utilized for the image transformation from low resolution to higher resolution to obtain more detailed information to identify the targets. The super-resolution has potential applications in various domains, such as medical image processing, crime investigation, remote sensing, and other image-processing application domains. The goal of the super-resolution is to obtain the image with minimal mean square error with improved perceptual quality. Therefore, this study introduces the perceptual loss minimization technique through efficient learning criteria. The proposed image reconstruction technique uses the image super-resolution generative adversarial network (ISRGAN), in which the learning of the discriminator in the ISRGAN is performed using the EfficientNet-v2 to obtain a better image quality. The proposed ISRGAN with the EfficientNet-v2 achieved a minimal loss of 0.02, 0.1, and 0.015 at the generator, discriminator, and self-supervised learning, respectively, with a batch size of 32. The minimal mean square error and mean absolute error are 0.001025 and 0.00225, and the maximal peak signal-to-noise ratio and structural similarity index measure obtained are 45.56985 and 0.9997, respectively.

Using Appetitive Motivation to Train Mice for Spatial Learning in the Barnes Maze.

The Barnes maze, a well-known spatial-learning paradigm, is based on the innate fear of rodents of large open spaces and their drive to hide. However, additional aversive stimuli (strong light and threatening sounds) are often necessary to provoke the hiding response while rendering the method cumbersome and more stressful. Our objective was to establish a Barnes maze-learning paradigm in mice using palatable food as a reward. After habituating male C57BL6/J or NMRI mice to the reward, the experimenter and the apparatus, either a slow (2 trials/day) or a massive conditioning schedule (4 trials/day), was run. Acquisition training was carried out until mice could locate the reward box with a maximum of one hole error. Then, the box was replaced to another location (reversal phase). Mice needed to relearn the new position with the same criterion. One week later, retention trials were performed. Both strains could reach the learning criteria; in the massive training within a shorter period. Spatial memory was demonstrated in the reversal and retention trials. Our results show that palatable food can be used as an efficient motivator to acquire allocentric navigation in the Barnes maze with the additional advantage of being less stressful.

Webtoon as an Attractive Learning Medium in Reading Comprehension for 6th Semester Students of English Education UIN Prof. K.H. Saifuddin Zuhri

This research is motivated by a situation where students' reading interest is low, so there is a need for media that can make students interested in reading and learning. Webtoon is an online comic application that is quite in demand by readers of various ages, Webtoons can also be used as learning media to increase motivation and reading comprehension. The aim of this research wants to discuss about how to use Webtoon as a medium for learning reading comprehension and whether Webtoon is in accordance with the computer-Assisted Language Learning Criteria (CALL). This study uses qualitative methods, data collected by interview techniques. The respondents in this study were 6 students from 6 TBI B who use the Webtoon application as a medium for learning reading comprehension. This research was analyzed using Chapelle theory, which has six criteria where indicate that an application can be learning media. The results of this study show that using Webtoon as a medium for learning reading comprehension is in accordance with the criteria for Computer-Assisted Language Learning (CALL), namely language learning potential, learner fit, meaning focus, authenticity, positive impact and practically. The majority of respondents gave positive comments on every aspect put forward by Chapelle. Therefore, Webtoons can be used as an interesting medium for learning reading comprehension.

Approaches to Language Teaching: Criticism from Theory to Practice

The criticism of any approach should be based on clear regulations. When we criticise any theory, our intention is that this criticism should meet some crucial requirement that should be systematic and not sporadic. “Systematic”, means that it is based on a specific set of principles or criteria, which are the same set of principles that are used in such a way as to evaluate all the existing teaching/learning theories. These principles are described as psychologically motivated principles, which are the richness of the input, the building of the students’ competency, and their involvement in meaningful communicative acts. The exhibition of an approach such as the Audiolingual methodology pushes us to investigate the question, what is problematic with it in the light of the learning criteria? One of the Audiolingual methodology’s apparent weaknesses is the output and the comprehensibility of the input. The Communicative Approach is good, but it is criticised because it does not provide a rich input. How can students communicate if the teachers fail to expose them to some linguistic stimulus? The Total Physical Response is known for the richness of the input, but the performance is poor. This paper is about learning how to criticise any language teaching approach and how to define a good one. The evaluation is not based on the students’ outcome, but on the benefit of finding out why a certain approach did not work in terms of what is taking place in the learners’ mind during the learning process.

EFEKTIVITAS PEMBELAJARAN MATEMATIKA MELALUI PENERAPAN MODEL COOPERATIVE INTEGRATED READING AND COMPOSITION PADA SISWA KELAS VII SMP MUHAMMADIYAH 5 MAKASSAR

This type of research is a pre-experimental research involving one class as an experimental class with the aim of knowing the effectiveness of mathematics learning through the application of the Cooperative model of Integrated Reading And Composition (CIRC) in class VII SMP Muhammadiyah 5 Makaasar. This study refers to the effectiveness criteria of learning, namely (1) Student mathematics learning outcomes with the application of the Cooperative model of Integrated Reading And Composition (CIRC). (2) Mathematical learning activities of students with the application of the Cooperative model of Integrated Reading And Composition (CIRC) model and (3) Students' responses to the application of the Cooperative model of Integrated Reading And Composition (CIRC). The research design is One-Group Pretest-Posttest design. The experimental sample was students of class VII.A SMP Muhammadiyah 5 Makassar. Data collection techniques used were test results of learning, observation sheets of student activities, observation of teacher activities and student questionnaire responses. Data analysis techniques used are descriptive analysis and inferential using the t-test. The results of the study indicate that after going through the application of the cooperative model of integrated reading and composition (CIRC) descriptive analysis results show that: (1) The average score of the initial test (pretest) the results of students mathematics learning was 28,84 with a standard deviation of 12,46 and the average score of the final test (postest) was 82,04 with a standard deviation of 5,704. From these results showed that 23 students have achieved mastery of the individual and this means that the mastery of the classical has been reached. (2) The average percentage of students achieving the expected frequenci of activity effectively, ie 80,50. (3) Questionnaire student responses indicate that students response to the cooperative model of integrated reading and composition (CIRC) which is 85,2 %. (4) The ability of teachers to manage learning mathematics through of the cooperative model of integrated reading and composition (CIRC) are in the very good category ie 3,58 %. Inferential analysis results indicate that the data pretest and postest has met normality test with p = 0,05. The results of testing the hypothesis by using t-test one sample test with previously made Normalized gain on the data pretest and postest data is obtained P = 0,000 = 0,05. So that rejected and be accepted. From the results of the above research, it can be concluded that mathematics learning through the application of the of the cooperative model of integrated reading and composition (CIRC) is effectively applied to class VII students. SMP Muhammadiyah 5 Makassar.

Generalized minimum error entropy for robust learning

The applications of error entropy (EE) are sometimes limited because its shape cannot be flexibly adjusted by the default Gaussian kernel function to adapt to noise variation and thus lowers the performance of algorithms based on minimum error entropy (MEE) criterion. In this paper, a generalized EE (GEE) is proposed by introducing the generalized Gaussian density (GGD) as its kernel function to improve the robustness of EE. In addition, GEE can be further improved to reduce its computational load by the quantized GEE (QGEE). Furthermore, two learning criteria, called generalized minimum error entropy (GMEE) and quantized generalized minimum error entropy (QGMEE), are developed based on GEE and QGEE, and new adaptive filtering (AF), kernel recursive least squares (KRLS), and multilayer perceptron (MLP) based on the proposed criteria are presented. Several numerical simulations indicate that the performance of proposed algorithms performs better than that of algorithms based on MEE.

SVM ensemble training for imbalanced data classification using multi-objective optimization techniques

One of the main problems with classifier training for imbalanced data is defining the correct learning criterion. On the one hand, we want the minority class to be correctly recognized, and on the other hand, we do not want to make too many mistakes in the majority class. Commonly used metrics focus either on the predictive quality of the distinguished class or propose an aggregation of simple metrics. The aggregate metrics, such as Gmean or AUC, are primarily ambiguous, i.e., they do not indicate the specific values of errors made on the minority or majority class. Additionally, improper use of aggregate metrics results in solutions selected with their help that may favor the majority class. The authors realize that a solution to this problem is using overall risk. However, this requires knowledge of the costs associated with errors made between classes, which is often unavailable. Hence, this paper will propose the semoos algorithm - an approach based on multi-objective optimization that optimizes criteria related to the prediction quality of both minority and majority classes. semoos returns a pool of non-dominated solutions from which the user can choose the model that best suits him. Automatic solution selection formulas with a so-called Pareto front have also been proposed to compare state-of-the-art methods. The proposed approach will train a svm classifier ensemble dedicated to the imbalanced data classification task. The experimental evaluations carried out on a large number of benchmark datasets confirm its usefulness.

The whereabouts of interorganizational learning: a maritime case study

Purpose This paper aims to present the results from a case study that investigated interorganizational learning in a buyer and seller relationship in the context of the maritime industry. This examination emphasized unraveling how the buyer and seller in the case study interacted and transferred knowledge when using a new business model that relied on servitization. Furthermore, this paper also addresses and discusses work practices, and the relationship between intra- and interorganizational learning. Design/methodology/approach A case study entailing the introduction of digital technology and a new business model into the maritime industry was used as an empirical example of interorganizational learning. The case study was conducted over a period of over one year and focused on a buyer of freight ships and a seller of servitized technology used on the ships. The organizations involved were the ships, the shipowner’s office and the ship engine supplier. The primary data acquisition methods comprised semi-structured interviews and observations. Findings The case identified interorganizational learning within the organizations at the individual, group and organizational levels, but only a few learning signs could be viewed as bidirectional interorganizational learning that can create knowledge and competitive advantages for the organizations. This is explained by the interorganizational learning context and the organizations’ motivation for learning at a strategic level. Originality/value This paper addresses an identified need for empirical studies on how interorganizational learning unfolds within organizations and connects to intraorganizational learning. Interorganizational learning studies often examine partnerships and joint ventures, in which partners have entered into these relationships with learning as a specific goal. By choosing a case in which interorganizational collaboration is anchored in operational matters, the study demonstrates the importance of motivation and agenda when entering into partnerships, concerning how inter- and intraorganizational learning develops within organizations. Furthermore, approaching these levels from an interrelated and practice-oriented perspective challenges established success criteria for interorganizational learning.

Generalized minimum error entropy Kalman filter for non-Gaussian noise

Error entropy is a well-known learning criterion in information theoretic learning (ITL), and it has already been applied to a wide range of fields. However, the shape of error entropy cannot be changed freely since its kernel function is the Gaussian kernel function, which causes the error entropy-based algorithm to handle only some specific kinds of noises. Benefiting from the property that the generalized Gaussian kernel function is free to adjust its shape, a novel Kalman-type filter algorithm based on the generalized minimum error entropy (GMEEKF) criterion is derived. Moreover, the mean error behavior, mean square error behavior, and computational complexity of the GMEEKF algorithm are analyzed. Finally, several simulations and experiments are performed to demonstrate the performance of the GMEEKF algorithm in comparison with the existing Kalman-type filter algorithms.

Signal Detection in MIMO Communications System with Non-Gaussian Noises based on Deep Learning and Maximum Correntropy Criterion

In this paper, we study signal detection in multi-input-multi output (MIMO) communications system with non-Gaussian noises such as Middleton Class A noise, Gaussian mixtures and alpha stable distributions, using several deep neural network-based detector models such as FULLYCONNECTED and DETNET detector. By applying information theoretic criterion of Maximum Correntropy , SVD analysis on the channel matrix and reducing network complexity, the suggested deep neural network detector performs well in environments with non-Gaussian noises and, compared to the deep neural network-based detector with MSE loss function, achieves better performance.

Spacing Effects in Task Repetition Research

AbstractThis article provides a conceptual review of the principles of input spacing as they might relate specifically to oral task repetition research and presents some of the common methodological considerations from the broader input spacing literature. The specific considerations discussed include the interaction between intersession intervals and retention intervals, the manipulation of posttests as a between‐participants variable, the number of task repetitions, absolute versus relative spacing, the criterion of learning, task type versus exact task repetition, and blocked versus interleaved practice. Each of these considerations is discussed with links, as appropriate, to the relevant empirical input spacing and task repetition literature. The purpose of this review is to highlight how, in many cases, these methodological considerations have been overlooked by task repetition researchers, including in studies where input spacing has and has not been a direct focus, and to suggest ways of addressing these methodological shortcomings in future research.