This study investigates the role of financial access in modulating the effect of education and lifelong learning on inequality in 48 African countries for the period 1996-2014. Lifelong learning is ...conceived and measured as the combined knowledge gained from primary through tertiary education while the three educational indicators are: primary school enrolment; secondary school enrolment and tertiary school enrolment. Financial development dynamics are measured with financial system deposits (liquid liabilities), financial system activity (credit) and financial system efficiency (deposits/credit). Three measures of inequality are employed notably: the Gini coefficient; the Atkinson index and the Palma ratio. The estimation strategy is based on the generalised method of moments. The following findings are established. First, primary school enrolment interacts with all financial channels to exert negative effects on the Gini index. Second, lifelong learning has negative net effects on the Gini index through financial deposit and efficiency channels. Third, for the most part, the other educational levels do not significantly influence inequality through financial access channels. Policy implications are discussed.
Can a new instructional approach influence lifelong learning and the development of competent lifelong learners? Blended and online learning provides a platform for learning that introduces ...technological affordance to enable learning. We seek to find an intersection between blended and online learning and lifelong learning through an instructional approach that encourages learners towards management of their own learning. This opens the door to becoming an autonomous, capable, self‐directed lifelong learner. In this context, heutagogy offers an instructional approach that may connect blended and online learning settings with the development of lifelong learning competence. After conducting a systematic literature review using the terms heutagogy, blended and online learning, and lifelong learning, literature that considers how to inspire and build human agency capabilities over the lifespan was chosen for Delphi method expert review. Using this methodology, we explore the possibility that online and blended higher education will contribute, where heutagogical experiences exist, to technology‐enabled lifelong learning. Results corroborate the idea that heutagogy and lifelong learning are intertwined by some common principles and that these are applicable to both blended and online learning settings and lifelong learning.
Practitioner notes
What is already known about this topic
Recent, and what is often continuous, change is impacting all we do, including the design and delivery of education.
This change requires new instructional models that improve immediate learning outcomes and prepares learners for learning across the lifespan.
The use of instructional processes labeled heutagogy include the opportunity for, and application of, activities of learning self‐direction, ‐determination, and ‐regulation, which can be helpful, even essential, for lifelong learning.
What this paper adds
This paper identifies an informed perspective, from data, that heutagogical design must be consciously implemented and supported for online and blended learning by instructional designers, instructors, and institutional leadership and infrastructure.
It is reasonable to suggest that online and blended learning could contribute, where heutagogical learning opportunities exist, to technology‐enabled lifelong learning.
Instructional practices that include choice, flexible or negotiated assessment, facilitation of reflection, learner confidence development, and involvement of the learner in designing their learning can be considered heutagogical.
Implications for practice and/or policy
Develop policy in support of a change in instructional practice that embraces a heutagogical approach in the design of courses to foster greater self‐directed and lifelong learning.
Educational development to support instructors to understand heutagogy and how it can be applied in the design and delivery of blended and online learning to foster technology enabled lifelong learning.
With the implementation of a heutagogical approach, student orientation along with purposeful scaffolding needs to be implemented to support students as they become more autonomous learners in technology‐enabled settings.
For future learning systems, incremental learning is desirable because it allows for: efficient resource usage by eliminating the need to retrain from scratch at the arrival of new data; reduced ...memory usage by preventing or limiting the amount of data required to be stored - also important when privacy limitations are imposed; and learning that more closely resembles human learning. The main challenge for incremental learning is catastrophic forgetting, which refers to the precipitous drop in performance on previously learned tasks after learning a new one. Incremental learning of deep neural networks has seen explosive growth in recent years. Initial work focused on task-incremental learning, where a task-ID is provided at inference time. Recently, we have seen a shift towards class-incremental learning where the learner must discriminate at inference time between all classes seen in previous tasks without recourse to a task-ID. In this paper, we provide a complete survey of existing class-incremental learning methods for image classification, and in particular, we perform an extensive experimental evaluation on thirteen class-incremental methods. We consider several new experimental scenarios, including a comparison of class-incremental methods on multiple large-scale image classification datasets, an investigation into small and large domain shifts, and a comparison of various network architectures.
As a documentary institution, museums within tertiary institutions have the potential to support and be supported by various scientific fields. This study aims to map the potential of museums in line ...with the Independent Learning Independent Campus (MBKM) and find out scientific collaborations that can strengthen the existence of museums in tertiary institutions in the context of MBKM. This study used a qualitative approach, using data collection techniques in the form of observation, literature review, and in-depth interviews with three supporting informants and one key informant. The study found that the higher education museum has great potential and fulfills all the requirements as an MBKM partner. Museums can be partners for internships, entrepreneurial projects, technology projects, social engineering applications, and even humanitarian projects. Museums can collaborate in informal educational activities, exchange of experts, teaching practitioners, research, and community service, all of which can involve lecturers and students from various fields of science.
In today's organisations and politics, there is a growing awareness of the gap between existing and needed digital competencies of the workforce to master the challenges of the digitalised future at ...work. Nevertheless, no comprehensive framework or definition of digital competencies at work has been proposed so far. Our aim is to offer a holistic view and broaden the scope of the concept of digital competencies, thereby focussing on applications at work. We combine diverse methods to integrate different perspectives on digital competencies. By conducting an extensive literature review about definitions and frameworks of digital competencies that might be applicable at work, we provide an overview of the current state of the art in research on digital competencies. Additionally, eleven half-structured interviews based on the critical incidents technique (CIT) were conducted to gain insights into the perspectives of professionals with expertise in digitalisation processes and digital competencies. Subsequently, researchers with different educational backgrounds clustered the results from both approaches and agreed on twenty-five dimensions that constitute digital competencies for white-collar workers with office jobs, encompassing a large variety of knowledge, skills, and abilities. The results of this research indicate that even though there is overlapping content, each perspective adds unique content to the concept of digital competencies at work. By creating a coherent and detailed framework and a definition, our research enhances the applicability of professional learning and development of digital competencies at work.
•Integrate perspectives from science and practice on digital competencies at work.•Combine findings from extensive literature review and semi-structured interviews.•Postulate a definition of digital competencies at work to foster understanding.•Propose comprehensive framework with 25 dimensions of digital competencies at work.
One side-effect of the COVID-19 pandemic has been increased enrollment in online classes. The paper explores the surge in activity from March through June 2020 in two massive open online classes ...(MOOCs) on Astronomy, offered by Coursera and Udemy. The increase in enrollment in both classes was an order of magnitude over the similar time span in previous years. Learners enrolling during the pandemic were more likely to be younger than thirty and less likely to have advanced degrees. A majority were full-time undergraduate students, and relatively few were professionals working in technical fields. The largest number of new students were from India and overall, the biggest surge in enrollment came from people in developing countries, particularly in Asia. Those who enrolled during the pandemic were more likely to take the course to get a certificate or to further their career goals than because they had intrinsic interest in the subject. Social motivations were important, particularly among full-time students in the course. These results, albeit limited to MOOCs in astronomy, suggest that new audiences have been turning to online classes during the pandemic for gaining credentials or advancing their professional skills.
Continuing technical vocational education and training (TVET) and professional development of the entire automotive sector towards implementing Industry 4.0 and 5.0 tools, methods and technologies ...has become an essential component of today's knowledge society. The principle of lifelong learning permeates all areas of professional and private life and has manifested itself as an essential component of today's knowledge society. Especially the automotive sector is challenged to successfully master the green and digital transformation. This needs engineers with expertise to actively shape the dual transformation. In this context, continuing education and training on a postgraduate level is indispensable. This paper illustrates continuing technical vocational education and training at Graz University of Technology, managed by the Life Long Learning organizational unit. As an example, a course concept is described in more detail on the basis of the university course Automotive Mechatronics. The aim of this paper is to present a successful TVET program and to show how postgraduate education over the lifespan can succeed sustainably for engineering professionals in the automotive sector.
Question Answering (QA) systems have witnessed a significant advance in the last years due to the development of neural architectures employing pre-trained large models like BERT. However, once the ...QA model is fine-tuned for a task (e.g., a particular type of questions over a particular domain), system performance drops when new tasks are added along time, (e.g., new types of questions or new domains). Therefore, the system requires a retraining but, since the data distribution has shifted away from the previous learning, performance over previous tasks drops significantly. Hence, we need strategies to make our systems resistant to the passage of time. Lifelong Learning (LL) aims to study how systems can take advantage of the previous learning and the knowledge acquired to maintain or improve performance over time. In this article, we explore a scenario where the same LL based QA system suffers along time several shifts in the data distribution, represented as the addition of new different QA datasets. In this setup, the following research questions arise: (i) How LL based QA systems can benefit from previously learned tasks? (ii) Is there any strategy general enough to maintain or improve the performance over time when new tasks are added? and finally, (iii) How to detect a lack of knowledge that impedes the answering of questions and must trigger a new learning process? To answer these questions, we systematically try all possible training sequences over three well known QA datasets. Our results show how the learning of a new dataset is sensitive to previous training sequences and that we can find a strategy general enough to avoid the combinatorial explosion of testing all possible training sequences. Thus, when a new dataset is added to the system, the best way to retrain the system without dropping performance over the previous datasets is to randomly merge the new training material with the previous one.
•Question Answering systems suffer over time several shifts in data distribution.•Performance after learning a new dataset is sensitive to previous training sequences.•We find a strategy to avoid the need of testing all possible training sequences.•Random merge of new training material with previous one avoids performance drop.
It was recently shown that architectural, regularization and rehearsal strategies can be used to train deep models sequentially on a number of disjoint tasks without forgetting previously acquired ...knowledge. However, these strategies are still unsatisfactory if the tasks are not disjoint but constitute a single incremental task (e.g., class-incremental learning). In this paper we point out the differences between multi-task and single-incremental-task scenarios and show that well-known approaches such as LWF, EWC and SI are not ideal for incremental task scenarios. A new approach, denoted as AR1, combining architectural and regularization strategies is then specifically proposed. AR1 overhead (in terms of memory and computation) is very small thus making it suitable for online learning. When tested on CORe50 and iCIFAR-100, AR1 outperformed existing regularization strategies by a good margin.
As a special case of machine learning, incremental learning can acquire useful knowledge from incoming data continuously while it does not need to access the original data. It is expected to have the ...ability of memorization and it is regarded as one of the ultimate goals of artificial intelligence technology. However, incremental learning remains a long term challenge. Modern deep neural network models achieve outstanding performance on stationary data distributions with batch training. This restriction leads to catastrophic forgetting for incremental learning scenarios since the distribution of incoming data is unknown and has a highly different probability from the old data. Therefore, a model must be both plastic to acquire new knowledge and stable to consolidate existing knowledge. This review aims to draw a systematic review of the state of the art of incremental learning methods. Published reports are selected from Web of Science, IEEEXplore, and DBLP databases up to May 2020. Each paper is reviewed according to the types: architectural strategy, regularization strategy and rehearsal and pseudo-rehearsal strategy. We compare and discuss different methods. Moreover, the development trend and research focus are given. It is concluded that incremental learning is still a hot research area and will be for a long period. More attention should be paid to the exploration of both biological systems and computational models.
PDF
