Search Results (12)

Federated learning (FL) enables collaborative model building among a large number of participants without sharing sensitive data to the central server. Because of its distributed nature, FL has limited control over local data and the corresponding training process. Therefore, it is susceptible to data poisoning attacks where malicious workers use malicious training data to train the model. Furthermore, attackers on the worker side can easily manipulate local data by swapping the labels of training instances, adding noise to training instances, and adding out-of-distribution training instances in the local data to initiate data poisoning attacks. And local workers under such attacks carry incorrect information to the server, poison the global model, and cause misclassifications. So, the prevention and detection of such data poisoning attacks is crucial to build a robust federated training framework. To address this, we propose a prevention strategy in federated learning, namely confident federated learning, to protect workers from such data poisoning attacks. Our proposed prevention strategy at first validates the label quality of local training samples by characterizing and identifying label errors in the local training data, and then excludes the detected mislabeled samples from the local training. To this aim, we experiment with our proposed approach on both the image and audio domains, and our experimental results validated the robustness of our proposed confident federated learning in preventing the data poisoning attacks. Our proposed method can successfully detect the mislabeled training samples with above 85% accuracy and exclude those detected samples from the training set to prevent data poisoning attacks on the local workers. However, our prevention strategy can successfully prevent the attack locally in the presence of a certain percentage of poisonous samples. Beyond that percentage, the prevention strategy may not be effective in preventing attacks. In such cases, detection of the attacked workers is needed. So, in addition to the prevention strategy, we propose a novel detection strategy in the federated learning framework to detect the malicious workers under attack. We propose to create a class-wise cluster representation for every participating worker by utilizing the neuron activation maps of local models and analyze the resulting clusters to filter out the workers under attack before model aggregation. We experimentally demonstrated the efficacy of our proposed detection strategy in detecting workers affected by data poisoning attacks, along with the attack types, e.g., label-flipping or dirty labeling. In addition, our experimental results suggest that the global model could not converge even after a large number of training rounds in the presence of malicious workers, whereas after detecting the malicious workers with our proposed detection method and discarding them from model aggregation, we ensured that the global model achieved convergence within very few training rounds. Furthermore, our proposed approach stays robust under different data distributions and model sizes and does not require prior knowledge about the number of attackers in the system. Full article

This study investigates the impact of flipped learning (FL) with and without teacher feedback on resilience and perceived professional competence among university students. This quasi-experimental study, conducted over three academic years, involved 255 s-year students (mean age = 20.47 ± 2.63; 60.4% female) enrolled in a Primary Education Teaching degree program at a Spanish public university. Participants were divided into two treatment groups: Feedback FL and Non-feedback FL. Validated scales—the Brief Resilience Scale (BRS) and the Perceived Competence Scale for Students (PCNS)—were used for pre- and post-intervention assessments. Both groups demonstrated significant improvements in resilience and perceived competence, with no statistically significant differences between the Feedback FL and Non-feedback FL groups. Data analysis included Mann–Whitney U tests for inter-group comparisons, Wilcoxon tests for intra-group comparisons, and 2 × 2 repeated measures ANOVA, which revealed no interaction effects (Time × Treatment) for either variable. These results suggest that FL alone fosters the development of transversal skills such as resilience and perceived competence, regardless of teacher feedback. The findings highlight the effectiveness of active learning models like FL in higher education for promoting essential competencies. Future research should address additional soft skills and employ larger, more diverse samples to further explore the role of feedback and innovative methodologies in enhancing FL’s impact. Full article

Background: Flipped learning (FL) is being considered, in terms of new educational trends, a beneficial pedagogical model in the classroom. In particular, FL and intrinsic motivation (IM) are key components to the model since they can be crucial to a high-quality education. FL for the development of IM in higher education, as well as searches for potential interventions have, thus improved over the past ten years. However, no reviews that analyze the findings and conclusions reached have been published. Consequently, the objectives of this paper were to analyze the relationship between the use of FL and the IM of students in higher education, and to identify the aspects that should be present in FL models to develop the IM that contributes to high-quality education. Methods: in accordance with PRISMA guidelines, a systematic review of PubMed, Web of Science, Scopus, and ProQuest was carried out. Results: Of the 407 studies that were initially discovered, 17 underwent a full examination in which all findings and conclusions were analyzed. After implementation, the majority of the FL interventions improved IM results. Conclusion: many key aspects have been identified that must be followed in order to intrinsically motivate students using the FL methodology. Full article

Federated learning (FL) is an emerging machine learning technique where machine learning models are trained in a decentralized manner. The main advantage of this approach is the data privacy it provides because the data are not processed in a centralized device. Moreover, the local client models are aggregated on a server, resulting in a global model that has accumulated knowledge from all the different clients. This approach, however, is vulnerable to attacks because clients can be malicious or malicious actors may interfere within the network. In the first case, these types of attacks may refer to data or model poisoning attacks where the data or model parameters, respectively, may be altered. In this paper, we investigate the data poisoning attacks and, more specifically, the label-flipping case within a federated learning system. For an image classification task, we introduce two variants of data poisoning attacks, namely model degradation and targeted label attacks. These attacks are based on synthetic images generated by a generative adversarial network (GAN). This network is trained jointly by the malicious clients using a concatenated malicious dataset. Due to dataset sample limitations, the architecture and learning procedure of the GAN are adjusted accordingly. Through the experiments, we demonstrate that these types of attacks are effective in achieving their task and managing to fool common federated defenses (stealth). We also propose a mechanism to mitigate these attacks based on clean label training on the server side. In more detail, we see that the model degradation attack causes an accuracy degradation of up to 25%, while common defenses can only alleviate this for a percentage of ∼5%. Similarly, the targeted label attack results in a misclassification of 56% compared to 2.5% when no attack takes place. Moreover, our proposed defense mechanism is able to mitigate these attacks. Full article

Over the last few years, there have been several attempts to provide software tools for the development of federated learning (FL) models. However, both the complexity of the concept itself and the high entry barrier of these tools have meant that their adoption has been limited. Considering the related benefits, especially in terms of preserving data privacy, and the need for this type of solution in specific areas where data sharing is impossible, not only from a practical point of view but also from a legal and even ethical perspective, it is necessary to advance in solutions that allow its use to be democratised and its deployment to be extended. With this objective in mind, FLIP (Federated Learning Interactive Platform) has been developed as a comprehensive, easy-to-use fully functional web-based FL network management platform that eases and accelerates the usage of federated datasets by researchers in real scenarios. In this sense, FLIP has achieved a SUS score of 84.64, confirming a high level of perceived usability as expected. Taking this into account, FLIP can help increase the productivity and adoption of FL by a wider audience. Full article

The aim of the present study was to compare the results of two educational platforms for the development of Flipped Learning (FL) in the preclinical practices of paediatric dentistry: Edpuzzle and Moodle 3.4. Methods: Fifty students filled out a questionnaire on knowledge of pulp treatments in primary dentition (Pre-Q). They were divided into two groups: one watched a video on the pulpotomy technique before preclinical practice using Moodle 3.4 (group A) and the other used Edpuzzle (group B). On the day of practice, the students resolved any doubts with the teacher. Next, they performed a pulpotomy on an artificial tooth and answered the questionnaire again (Post-Q) together with a satisfaction survey. Results: In both groups, an increase in the number of correct answers was found in the Post-Q compared to the Pre-Q (p < 0.001), but with no significant differences between groups. The pulpotomy practice ratings were higher in group B, which used Edpuzzle (p < 0.001). In the satisfaction survey, we only found significant differences (p = 0.003) in access to the video, since 100% of the students in group A found it easy to view it through Moodle, unlike in group B. Conclusions: Our results suggest that the use of educational platforms specially designed for FL, such as Edpuzzle, can improve the qualifications of students in paediatric dentistry practices. Full article

Technology development has been integrated into the educational environment and has led teachers to become much better trained in educational, technological tools. Currently, education is being transformed; for this, there are new methodological approaches, and education needs the integration of digital tools. Previously and still today, traditional and not very creative strategies are applied in the teaching–learning process, which does not fully contribute to the progress of education. The present work focuses on using Information and Communication Technologies (ICT), considering the Flipped Learning Model (FL), an active methodology. However, it is also essential to know the appropriate ICT to apply during the learning process. The Information and Communication Technologies articulated with the Flipped Learning Model benefits and motivates students; in this way, through collaborative learning, communication between classmates and teachers is favored; in addition, it encourages autonomous work, helps the analysis of the contents in each of the subjects, and favors the construction of new knowledge. Therefore, knowing which ICTs are incorporated as the most efficient in the Flipped Learning Model is necessary. In this way, it is required to obtain information about the ICTs teachers have preferred to apply within the Flipped Learning Model and which ones are recommended from the classroom experience. The work shows which ICTs are most used and which ones benefit students to obtain significant learning. Consequently, considering the application of ICT and Flipped Learning in educational communities is a way to innovate the teaching–learning process. Full article

This article describes an empirical study to evaluate how the flipped learning (FL) approach has impacted a learner’s perception in attaining the graduate attributes (GAs) of five capstone project units offered at Melbourne Institute of Technology in Australia, where the authors are affiliated. The subjects include one undergraduate and one postgraduate business unit, and one undergraduate and two postgraduate units in networking. Our study is distinguished from previous research in two novel aspects. First, the subject matter concerns capstone project units which are taken by students in the final year of their degree. In these units, students are expected to apply a variety of knowledge and skills that they have acquired thus far in carrying out an industry-based project of substantial complexity. The learning outcomes (LOs) require students to apply skills and knowledge that they have learned across completed units and connect them with real-world problems. Second, the FL approach has been applied wholly in an online virtual classroom setting due to the social distancing restrictions enforced by local authorities in response to the COVID-19 pandemic. Our hypothesis is that FL has positively influenced the perception of learners in their attaining the GAs. We tested this hypothesis by using data collected by an online survey administered to the student cohorts of the five chosen units at the end of Trimester 1 of 2021. The survey, which comprised 14 questions, assesses a student’s perception of achieving the LOs through developments in three dimensions, including cognitive, affective, and behavioural, acquired in a real-world client setting. Statistical analyses of the survey data reveal that the FL approach resulted in a positive perception by students of their attaining the GAs through achieving the LOs of the capstone project units, which in turn is supported by the responses to the three measured dimensions. Full article

Nowadays, several institutions and universities have employed Flipped and Blended Learning models in the educational context. Consequently, these models advance correctly in the classrooms; students become more able to build their knowledge and develop their skills with technological innovation that could be compared with studies based on technological achievements and accessible data Tics. Therefore, this article proposes educational innovation in the evaluation processes within the FL and BL learning models to ensure active and interactive education. In this work, a bibliometric analysis of 2000 scientific articles from the Web of Science and Scopus databases has been carried out, thus making it possible to identify countries and universities with constant and relevant use of these learning models, as well as a comparison with these databases of articles, citations and their global impact. On the other hand, a survey was conducted with 41 teachers of different subjects and 385 students from elementary school to high school using the Likert scale. The questions were represented in perception values with their respective analysis of results and evaluations about these learning models, technological resources, flexible curriculum, learning before class, autonomous learning, and future professional competencies, which will help us know the existing motivation between them. Full article

The present article focuses on the role that the artificial teaching and learning of mathematics could play for education in the forthcoming era of a new industrial revolution that will be characterized by the development of an advanced Internet of things and energy, and by the cyber-physical systems controlled through it. Starting with a brief review of the traditional learning theories and methods of teaching mathematics, the article continues by studying the use of computers and of applications of artificial intelligence (AI) in mathematics education. The advantages and disadvantages of artificial with respect to traditional learning in the classroom are also discussed, and the article closes with the general conclusions and a few comments on the perspectives of future research on the subject. Full article

The purpose of the present study was to examine if and how the introduction of a flipped learning (FL) framework in Norwegian lower and upper secondary school physical education (PE) could affect student situational motivation and health-related fitness knowledge (HRFK). 206 Norwegian students (48% girls) from secondary and upper secondary schools were included in a three-week-long intervention. 85 students were assigned to an intervention group and 121 students to a control group, with the intervention aimed at integrating information communication technology (ICT) through the use of online videos. The results showed a negative motivational change for male students unless the activity changes were placed within an explanatory rationale through FL. The application of FL caused more cognitive knowledge learning, resulting in higher levels of HRFK among both girls and boys. The current research provided valuable insights into Norwegian students’ motivation for PE and that cognitive knowledge learning can be affected when integrating the use of ICT in PE. Full article

The evolution of technology in the educational field has resulted in the emergence of new ways of teaching and learning, adapted to the formative characteristics of a digital era. This is the case of Flipped Learning (FL), an innovative teaching methodology that uses technology to carry out instructional action and change learning moments and spaces. The objective of this study is to verify the effectiveness of FL on a traditional methodology without the use of digital resources in teaching content related to intercultural education. A quantitative research method has been followed by an experimental design. Two study groups have been set up. With the control group a traditional training methodology has been used, and with the experimental group an innovative one, through FL. A total of 60 students of the 4th level of Secondary Education from an educational center in Ceuta (Spain) have participated. An ad hoc questionnaire derived from other validated instruments has been used for data collection. The findings show that the FL has achieved better results in the different dimensions analyzed. Therefore, it is concluded that the implementation of innovative training actions such as FL can obtain better results in academic indicators than a traditional methodology. Full article