Active Learning Techniques Research Articles

Model update for intrusion detection: Analyzing the performance of delayed labeling and active learning strategies

Intrusion Detection Systems (IDS) help protect computer networks by identifying and detecting attempts to obtain unauthorized access to data via computer networks by inspecting packets separately or in the context of flows. Intrusion detection is a classification task performed on continuously generated packets with a non-stationary distribution (data stream). As such, the decision models used for Intrusion Detection must be constantly updated to account for changes in what constitutes normal traffic of a network and correctly identify attacks. This work evaluates two approaches concerning label availability to update the intrusion detection classifier. First, we analyze the impact of delayed labeling of samples on the classifiers' performance, and second, we evaluate the impact of using active learning strategies on the classifiers. Our experimental evaluation uses two datasets (CIC-IDS2017 and CSE-CIC-IDS2018) to compare different data stream classification algorithms under different evaluation measures. Based on comparison results, we studied different active learning techniques to estimate the impact of delayed labeling on packet-based IDS. We found that the performance of the classifiers is inversely proportional to the label delivery rate. Besides, the active learning strategies helped keep the performance compatible with the baselines, even with a small set of labeled instances.

An active learning approach to train a deep learning algorithm for tumor segmentation from brain MR images

PurposeThis study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model.MethodsThe publicly available training dataset provided for the 2021 RSNA-ASNR-MICCAI Brain Tumor Segmentation (BraTS) Challenge was used in this study, consisting of 1251 multi-institutional, multi-parametric MR images. Post-contrast T1, T2, and T2 FLAIR images as well as ground truth manual segmentation were used as input for the model. The data were split into a training set of 1151 cases and testing set of 100 cases, with the testing set remaining constant throughout. Deep convolutional neural network segmentation models were trained using the NiftyNet platform. To test the viability of active learning in training a segmentation model, an initial reference model was trained using all 1151 training cases followed by two additional models using only 575 cases and 100 cases. The resulting predicted segmentations of these two additional models on the remaining training cases were then addended to the training dataset for additional training.ResultsIt was demonstrated that an active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas (0.906 reference Dice score vs 0.868 active learning Dice score) while only requiring manual annotation for 28.6% of the data.ConclusionThe active learning approach when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.Critical relevance statementActive learning concepts were applied to a deep learning-assisted segmentation of brain gliomas from MR images to assess their viability in reducing the required amount of manually annotated ground truth data in model training.Key points• This study focuses on assessing the performance of active learning techniques to train a brain MRI glioma segmentation model.• The active learning approach for manual segmentation can lead to comparable model performance for segmentation of brain gliomas.• Active learning when applied to model training can drastically reduce the time and labor spent on preparation of ground truth training data.Graphical

An innovative approach to comprehensive communication skills training for residents: a resident-led communication curriculum.

Innovations to enhance residency training in interpersonal and communication skills are needed and a resident-led strategy has not been well-described. In this study, we explored a resident-led comprehensive communication skills curriculum for internal medicine residents. Residents and faculty prepared the curriculum as part of an Accreditation Council for Graduate Medical Education (ACGME) Back to Bedside Project and with "The language of caring guide for physicians." Employing active learning techniques, three residents led 43 internal medicine residents in seven 1 h sessions from 2019 to 2020. Using a 35-question survey, we assessed pre and post self-reported competence in: mindful practice, collaboration and teamwork, effective openings and closing, communicating with empathy, effective explanations, engaging patients and families as partners, and hard conversations. A Wilcoxon signed rank test was employed to explore differences in median scores after matching each person's pretest and posttest score. The median score for aggregate communication and the scores for all seven competencies assessed improved from pre to post (p < 0.05). This indicates that residents reported higher incidences of performing patient-centered communication skills after the curriculum compared to before. Using a five-point Likert scale, 100% of participants agreed the program improved their communication skills and improved confidence in bedside patient-centered communications. A resident-led comprehensive communication skills curriculum for internal medicine residents was implemented showing improvement in skills over the course of the curriculum. The curriculum was well-accepted by post-survey evaluation and was feasible with motivated resident-leaders, use of an existing guide to communication, and reserved didactic time to implement the program.

Using Rubrics to Evaluate E-Learning Tools in Radiology Education

Recent trends in medical education with an emphasis on active learning strategies and blended learning techniques have resulted in a growing volume and utilization of online educational resources. Integration of online learning tools into medical curricula has been further necessitated during the COVID-19 pandemic. With access to abundant digital radiology education resources, it has become increasingly important for educators to be able to evaluate the efficacy of e-learning tools for use in radiology education. In this manuscript, the authors describe their successful search for a vetted method to evaluate e-learning tools in radiology education. The selected rubric was designed by educational developers supporting technology to be used as a formative tool in higher education. The rubric was applied in condensed and noncondensed formats to 2 existing popular highly subscribed radiology e-learning resources and results were displayed in narrative and visual formats. More widespread application of this rubric would be beneficial to the content creators and learners alike.

Increasing Cultural Awareness in the Advanced Nursing Practice Program Through Active Learning.

As student cultural diversity increases in higher education, educators must implement bias-free evaluations and active learning activities. The objective structured clinical examination (OSCE), an active learning strategy where simulated patients role-play specific clinical scenarios, can be part of the advanced nursing practice (ANP) educational program for either formative or summative competencies evaluation. This article describes active learning and theatrical techniques used in a grant-funded workshop to increase cultural awareness among ANP students, nurse educators, and simulated patients. The participants' increased understanding of their potential biases can create a psychologically safe and bias-free OSCE for culturally diverse students.

Future of the Lebanese Middle School Science Teaching Methods: The Importance of Teacher’s Shifting from Passive Learning to More Active Learning in Science Classes

This study evaluated the importance of integrating active learning techniques in to Lebanese middle school science classes. It examined the effect of this modification on students' academic performance and social development .The researcher collected data by using mixed method approach referred to as sequential explanatory design where the research collected both quantitative and qualitative data. The quantitative data was collected by distributing questionnaire developed based on 5 point Likert- scale to a non-probability convenience sample (N=50) of science teachers and the qualitative data was collected by performing an online interview with four middle school science teachers. The researcher used Statistical Package for the social sciences program (SPSS) to analyze quantitative data and to answer the studied hypothesis and problematic by using three tests (a) one sample T-test, (b) one way Manova test, and (c) correlation test. The interview data was analyzed qualitatively .The findings of this study showed that (a) there is a significant importance for teachers’ shifting from passive learning to more active learning in Lebanese middle school science classes, (b) Lebanese middle school science teachers are modifying their teaching methods in their classes to improve students' academic performance and social development, and (c) there is a significant effect of the teaching methods modification on students' academic performance and social development based on the Lebanese middle school teachers' perspective.

Designing formulations of bio-based, multicomponent epoxy resin systems via machine learning

Petroleum-based epoxy resins are commonly used as a matrix in fiber-reinforced polymer composites. Bio-based epoxy resin systems could be a more environmentally friendly alternative to conventional epoxy resins. In this work, novel formulations of multicomponent, amino acid-based resin systems exhibiting high or low glass-transition temperatures (Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document}) were designed via Bayesian optimization and active learning techniques. After only five high-Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document} experiments, thermosets with Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document} already higher than those of the individual components were obtained, pointing out the existence of synergistic effects among the amino acids used and confirming the efficiency of the theoretical design. Linear and nonlinear machine learning (ML) models successfully predicted Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document} with a mean absolute error of 3.98∘C\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{\\circ }{\ ext{C}}$$\\end{document} and R2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$R^2$$\\end{document} score of 0.91. A price reduction of up to 13.7% was achieved while maintaining the Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document} of 130∘C\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$^{\\circ }{\ ext{C}}$$\\end{document} using an optimized formulation. The LASSO model provided information about the dependence of Tg\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$T_{{\ ext{g}}}$$\\end{document} on the number of active hydrogen atoms and aromaticity. This study highlights the importance of Bayesian optimization and ML to achieve a more sustainable development of epoxy resin materials.Impact statementThis article shows how the sustainability of epoxy resin systems (ERSs) can be significantly improved by combining experimental and theoretical strategies. First, amino acids are used as curing agents in multicomponent formulations to produce bio-based ERSs. Second, the number of trial-and-error experiments required to obtain formulations with high or low glass-transition temperatures (Tg) is greatly reduced using machine learning (ML) strategies to design all experiments. Not only is it shown how Tg can be maximized in only five new theoretically designed formulations, but the economic advantages of the proposed approach are also discussed. The trends between Tg and the type of optimized biocomponents are discussed based on the unambiguous interpretation of the best-trained ML model. The results presented in this study pave the way for the theoretical design of more sustainable polymeric materials.Graphical abstract

The Support of History Subject in Improving Reading Literacy among Students at State Senior High School 3 Pematangsiantar

This research discusses the reading literacy mentoring for students at SMA Negeri 3 Pematangsiantar through the history subject. The low interest in reading among students in Indonesia can lead to stagnant educational quality. The mentoring program includes strategies such as integrating reading literacy into history lessons, providing relevant reading materials, using active learning techniques, involving the local community, mentoring students, and utilizing digital technology. Despite challenges such as limited resources and low reading interest, this mentoring program aims to improve students' reading skills and enhance their understanding of history.

Project CARPE Diem: Curriculum and Assessment for Residents on Pediatric Epilepsy

BackgroundPediatric residencies are not preparing trainees well to manage patients with seizures and epilepsy. To address this, we implemented a six-session curriculum using active learning techniques with the goals of improving the knowledge and attitudes needed to effectively identify and treat seizures and epilepsy, and increase residents’ comfort with counseling families about these topics. MethodsA structured curriculum was implemented over the course of one month for the pediatric resident learner using andragogical methods such as didactic presentation, small-group sessions, role play, and simulation. A 15 multiple-choice question (MCQ) assessment and 13-statement attitudes inventory (AI) using Likert scales were given before the start of the curriculum and after its completion. Pretest and post-test scores were compared for the residents who attended at least three of the six sessions. ResultsFifty residents completed the pretest, 34 residents completed the post-test, and 24 residents completed both. Of the 24, only 19 residents attended at least half the sessions. There was a significant improvement in scores for the MCQ assessment, for each grouped AI domain, and in the proportion of residents who reported comfort with 11 of 13 AI statements. ConclusionThe findings of this study demonstrate that given quality high-impact educational content, pediatric residents’ knowledge about seizures and epilepsy improves significantly. It is feasible to implement active learning techniques during structured didactic time. The importance of closing the knowledge gap and improving pediatric comfort with seizures and epilepsy is essential and may translate to the much needed improvement in access to care for the patients with these disorders.

Enhancing Motivation Strategies in a Mandatory University Geography Course

This study investigates the effectiveness of innovative teaching strategies designed to enhance student motivation and engagement in a compulsory Physical Geography of the Iberian Peninsula course taught to first-year history students at the University of Seville. The paper outlines the challenges faced in teaching large and diverse groups of students, and the need to develop instructional approaches that accommodate various learning preferences and academic backgrounds. The re-search evaluates the impact of the implemented strategies on students' perception of the subject matter, with a specific focus on the shift in rankings from the beginning to the end of the academic year. The findings demonstrate a significant improvement in students' appreciation for the subject, as evidenced by a decrease in the percentage of students who ranked the course as their least preferred (from 40.7% to 4.9%) and an increase in the percentage who ranked it as their most preferred (from 0% to 11.1%). The study concludes that the adoption of motivational strategies, including active learning techniques, collaborative tasks, and personalized feedback, can lead to substantial improvements in student engagement and overall learning outcomes in a compulsory university course. The implications for teaching practice and future research are discussed.

A unified microstructure segmentation approach via human-in-the-loop machine learning

Microstructure segmentation, a technique for extracting structural statistics from microscopy images, is an essential step for establishing quantitative structure–property relationships in a broad range of materials research areas. However, the task is challenging due to the morphological complexity and diversity of structural constituents. While recent breakthroughs in deep learning have led to significant progress in microstructure segmentation, there remain two fundamental challenges to its further diffusion: prohibitive annotation costs and an unreliable decision-making process. To tackle these challenges, we propose a human-in-the-loop machine learning framework for unified microstructure segmentation, which leverages recent advances in both weakly supervised learning and active learning techniques. The key idea behind our approach lies in the integration of human and machine capabilities to make not only precise but also reliable microstructure segmentation at minimal annotation costs. Extensive experiments demonstrate the generality of our framework across different material classes, structural constituents, and microscopic imaging modalities.

Using Role-Play during Teaching Demonstrations to Practice Active Learning and Classroom Management Techniques

We designed and implemented an innovative teacher learning activity within the context of a graduate course. Course-takers created and completed a teaching demonstration and their “students” (i.e., fellow course-takers) were each assigned a specific role to play (e.g., disengaged) in order to provide the “teachers” with opportunities to practice their classroom management skills. After completing the demonstrations, the course-takers completed an assessment of their own and their colleagues’ teaching effectiveness and their attitudes toward the role-playing activity. Overall, course-takers reported that they learned valuable teaching skills and generally enjoyed the role-playing activity. Such role-playing activities can safely provide developing teachers with opportunities to practice various pedagogical strategies, while simultaneously helping them sharpen their classroom management skills.

Assessing the Effect of Pedagogical Transition on Classroom Design for Tertiary Education: Perspectives of Teachers and Students

Active learning has been increasingly important in tertiary education in recent years due to its powerfully favourable impact on students’ learning attitudes and efficacy. Indeed, the way that a classroom is set up has a direct impact on how well students learn and how well teachers teach. The continuous evaluation of students’ learning performance is essential for guiding future classroom renovations and creating a cutting-edge learning environment for both students and teachers. The aims of this paper were to provide a better understanding of the latest development trend of learning mode preference in tertiary education and to investigate any underlying similarities and differences in the perceptions between teachers and students. To support both teaching and learning, an empirical questionnaire survey was conducted among teachers and students in Hong Kong to assess the effectiveness of various active learning techniques and passive learning techniques adopted in tertiary education. Opinion-based data were collected on the perceived benefits and disadvantages of both learning techniques as well as the importance of various classroom design features. To determine the significance of the variations in opinions between teachers and students on the survey responses, descriptive statistical analyses using the mean score and Mann–Whitney U-test were carried out. The results of the Mann–Whitney U-test on the advantages of traditional learning showed that the following variables significantly varied: ‘direct information from the teacher’; ‘timesaving (group discussion may waste time)’ and ‘allow more time for Q&amp;A’. These advantages were generally rated higher from the viewpoint of students rather than teachers. However, no significant difference was established concerning the limitations of traditional learning. The findings of this study can help teachers and instructors to understand how different teaching and learning methods affect students’ ability to learn effectively, which can ultimately help institutional policymakers to determine the necessary essential requirements for orchestrating classroom designs to create more conducive teaching and learning environments. The findings also aim to inform policymakers and educational institutions on the impact of pedagogical change on the fundamental design requirements for a flexible classroom environment supportive of students’ active learning, especially in tertiary education.

Toward a Diverse Next-Generation Energy Workforce: Teaching Artificial Photosynthesis and Electrochemistry in Elementary Schools through Active Learning

Artificial photosynthesis is a promising approach to generate commodity chemicals using abundant chemical feedstocks and renewable energy sources. Despite its importance, affordable and effective hands-on classroom activities that demonstrate artificial photosynthesis and teach key concepts, especially for primary school students, are lacking. Educating young students on this topic is a critical step in the development of the next-generation energy workforce, especially one that is diverse in race and gender. We hypothesize that an effective approach to educate a broad range of young students on the topic of artificial photosynthesis is through the use of an active learning-based lesson plan that employs cheap and accessible materials. This hypothesis is confirmed by evaluating the understanding of fifth grade students with a survey before and after a lesson plan on artificial photosynthesis that uses active-learning techniques and uses safe and highly accessible materials (baking soda, tap water, plastic jars, Ni coil, alligator clips, and a solar cell) to perform solar-powered water splitting. The lesson plan and survey questions are designed to align with the educational outcomes for fifth grade classrooms in California and to address four general learning objectives: (1) Motivations of Artificial Photosynthesis, (2) Applications of Artificial Photosynthesis, (3) Inputs and Outputs of Artificial Photosynthesis, and (4) Engineering Design for Artificial Photosynthesis. The survey data demonstrate a statistically significant improvement in overall student understanding from the lesson plan. Importantly, the data show that the lesson plan presented here is effective at narrowing the performance gap between minority students and overly represented groups.

The pedagogy of teacher education in Ethiopia: Reconstructing understandings and practices on teaching about teaching and learning to teach

There is an argument that Marshal McLuhan’s ‘the medium is the message’ adage is more likely to be underpinned in the context of teacher education since pedagogy is not only a means of delivering content in teacher education but also the content (what) in teachers’ professional preparation (Grossman, 2009). The quality of teacher education depends heavily on the practice of teacher educators and the learning experiences of student teachers: Pedagogy of teacher education. The pedagogy of teacher education is conceptualized as teaching about teaching and learning to teach. Using primary data from teacher educators and student teachers and secondary data sources (documents and research studies), the paper examined and reflected on Ethiopian teacher education pedagogy. Questionnaire, interview and observation were used to collect primary data. The findings revealed ingrained assumptions that student teachers learn to teach by learning theories and facts of different courses through lectures, discussions, and group work. Teacher educators in the study characterized their typical classroom as a combination of ‘reviewing of the previous lesson, explanation, discussion, and lesson summary’ which inherently reflects features of a typical classroom in Ethiopian schools. When asked which pedagogies help student teachers to learn to teach, teacher educators mentioned group work, discussion, lecturing, oral questioning and classwork, quiz, and practicum (ordered in terms of frequencies). It is argued that teaching about teaching and learning to teach are customarily reduced to the delivery and learning of course contents rather than the ‘how’ of teaching through teacher educators’ modeling and reflection, student teachers’ school experiences, or other pedagogies that develop learning to teach among student teachers. Teacher educators’ practices, however, in some cases demonstrated elements of active learning techniques aimed at improving mastery of content and practicing teaching skills. However, their interview responses on preparing teachers largely lack pedagogies relevant to learning to teach. Efforts to improve the quality of education in general and teacher education, in particular, should prioritize the critical examination and reform of teacher educators’ preparation, professional development of teacher educators, and pedagogical practices at teacher education colleges and universities.

Using instructor-developed study resources to increase evidence-based learning strategies among medical students: A mixed-methods study

Purpose Applying effective learning strategies to address knowledge gaps is a critical skill for lifelong learning, yet prior studies demonstrate that medical students use ineffective study habits. Methods To address this issue, the authors created and integrated study resources aligned with evidence-based learning strategies into a medical school course. Pre-/post-course surveys measured changes in students’ knowledge and use of evidence-based learning strategies. Eleven in-depth interviews subsequently explored the impact of the learning resources on students’ study habits. Results Of 139 students, 43 and 66 completed the pre- and post-course surveys, respectively. Students’ knowledge of evidence-based learning strategies was unchanged; however, median time spent using flashcards (15% to 50%, p < .001) and questions (10% to 20%, p = .0067) increased while time spent creating lecture notes (20% to 0%, p = .003) and re-reading notes (10% to 0%, p = .009) decreased. In interviews, students described four ways their habits changed: increased use of active learning techniques, decreased time spent creating learning resources, reviewing content multiple times throughout the course, and increased use of study techniques synthesizing course content. Conclusion Incorporating evidence-based study resources into the course increased students’ use of effective learning techniques, suggesting this may be more effective than simply teaching about evidence-based learning.

TEACHING BUSINESS ENGLISH: THE PROBLEM OF THE USE OF THE CASE METHOD

The article deals with the problem of usage of the case method in teaching business English is viewed as a crucial component of training for formation of business communication skills due to the ability to create and test actual circumstances that students can meet in business communication. The primary pedagogical benefits of employing case studies in corporate English classes are discussed in this study. The case method is an active learning technique that consists in the use of specific learning situations in the organization of the learning process, aimed at formulating a problem and finding options for solving it, followed by analysis in the classroom. Situational analysis is an active learning technique that calls for both instructor and student interaction in the classroom, in contrast to traditional education that relies on lectures and has little to no student involvement. The primary objective of the essay is to outline the benefits of the case study approach as a cutting-edge way for enhancing learners' linguistic and nonlinguistic skills.

Training for the United Nations in the Twenty-First Century; Professionalism Training on Leadership, Negotiation, and Gender for Model United Nations Simulations

Abstract The world needs people who understand the crucial importance of global governance. Training students to participate in a Model United Nations (MUN) simulation provides them with the critical thinking, negotiation, and diplomacy skills that prepare them to work in international relations positions and address negotiation and diplomacy challenges. This article provides a fresh look at why MUN is important by analyzing MUN training guides used to teach students how to participate in MUN simulations. It finds that the training guides offer excellent insight into the functions and structure of the United Nations, but little in the form of learning how to negotiate or engage in diplomacy. Specifically missing are discussions of leadership, negotiation from interest, not position (also known as principled negotiation), and how to train students to be inclusive and respond to discrimination or bias during negotiations. The article includes a description of active learning and learner-centered teaching techniques and provides insight into how a specific MUN team implemented professionalism training in the 2010s. The article is written in collaboration between an MUN advisor and students who trained as MUN delegates, integrating the voices and perspectives of those who train and those who are trained.

ACTIVE LEARNING AND ACADEMIC PERFORMANCE: THE CASE OF REAL-TIME INTERACTIVE STUDENT POLLING

Active-learning pedagogical practices are encouraged in the education literature to improve academic outcomes. The research reported in this paper explored the effect of real-time interactive polling—a type of active learning—on three measures of academic outcome: final marks, student engagement, and failure probability. To provide credible impact estimates, a cluster randomized controlled trial in an introductory statistics class was conducted. With a little over 500 students, nine out of the 24 tutorial classes were conducted with real-time quizzes; the other 15 tutorial classes served as control classes. The results showed that this active-learning technique did not impact final marks and the probability of failing the subject. It did, however, increase tutorial attendance by about 24%, which corresponded to about 1.6 additional hours compared to the control classes. The findings support the literature on the benefits of classroom response systems that increase student engagement, student interest, and enjoyment from smaller group discussions.

Machine learning interatomic potential for silicon-nitride (Si3N4) by active learning.

Silicon nitride (Si3N4) is an extensively used material in the automotive, aerospace, and semiconductor industries. However, its widespread use is in contrast to the scarce availability of reliable interatomic potentials that can be employed to study various aspects of this material on an atomistic scale, particularly its amorphous phase. In this work, we developed a machine learning interatomic potential, using an efficient active learning technique, combined with the Gaussian approximation potential (GAP) method. Our strategy is based on using an inexpensive empirical potential to generate an initial dataset of atomic configurations, for which energies and forces were recalculated with density functional theory (DFT); thereafter, a GAP was trained on these data and an iterative re-training algorithm was used to improve it by learning on-the-fly. When compared to DFT, our potential yielded a mean absolute error of 8 meV/atom in energy calculations for a variety of liquid and amorphous structures and a speed-up of molecular dynamics simulations by 3-4 orders of magnitude, while achieving a first-rate agreement with experimental results. Our potential is publicly available in an open-access repository.