The Pattern of Hybrid Learning to Maintain Learning Effectiveness at the Higher Education Level Post-COVID-19 Pandemic

<p style="text-align: justify;">Online learning during the Coronavirus disease (COVID-19) pandemic has awakened and affirmed the necessity of learning based on digital technology. The article was aimed to analyze the effectiveness of online learning at bachelor’s, master’s, and doctoral degrees of Islamic Religious Education as a reference to develop a learning pattern post-COVID-19 pandemic. The research employed a mixed-method design with a concurrent triangulation model. The samples were taken using stratified random and purposive sampling. Meanwhile, the data were collected through questionnaires, in-depth interviews, and forum group discussion. A descriptive analysis and one-way analysis of variance were used to analyze the quantitative data, while interpretative descriptive for the qualitative data. The research showed that online learning during the COVID-19 pandemic at the bachelor’s, master’s, and doctoral degrees of Islamic Religious Education have been effective. In detail, online learning at the doctoral degree was the most effective among all. On the other hand, face-to-face learning is still necessary. Therefore, the learning pattern developed post-COVID-19 pandemic combines face-to-face and online learning (hybrid learning). The formulation is adjusted to the characteristics, educational purpose and orientation, level of ability, readiness, and learning autonomy of the students at each educational level.</p>

GAYA BELAJAR MAHASISWA PADA MATA KULIAH BAHASA JEPANG BERBASIS ONLINE LEARNING

This research is predicated on the problem of learning that is faced by students in university during social distancing due to Coronavirus outbreak. One of the striking problems is most students are not comfortable with e-learning. The aim of this article is grasping the typology of students particularly in e-learning that is conjunction with four aspects namely learning patterns, doing assignments, learning habits and using tools. Furthermore, the other objective is the research will be contributed to lecturers, faculties, universities and policymakers in term of planning, creating, and developing e-learning systems. The methodology used in this research is quantitative with descriptive type where is the population or sample as the object of the research and it is only to describe a variable without comparing to other variables. The result of the study denotes the typology of visual learners dominating learning pattern and doing assignment which was 54% and 46% respectively. While in learning habit, the auditory type was the most popular which was 42% however the kinesthetic type had a low percentage because this activity was rarely carried out in e-learning hence that is why the quantity is negligible.

S$$^{2}$$ES: a stationary and scalable knowledge transfer approach for multiagent reinforcement learning

Knowledge transfer is widely adopted in accelerating multiagent reinforcement learning (MARL). To accelerate the learning speed of MARL for learning-from scratch agents, in this paper, we propose a Stationary and Scalable knowledge transfer approach based on Experience Sharing (S$$^{2}$$ 2 ES). The mainframe of our approach is structured into three components: what kind of experience, how to learn, and when to transfer. Specifically, we first design an augmented form of experience. By sharing (i.e., transmitting) the experience from one agent to its peers, the learning speed can be effectively enhanced with guaranteed scalability. A synchronized learning pattern is then adopted, which reduces the nonstationarity brought by experience replay, and at the same time retains data efficiency. Moreover, to avoid redundant transfer when the agents’ policies have converged, we further design two trigger conditions, one is modified Q value-based and another is normalized Shannon entropy-based, to determine when to conduct experience sharing. Empirical studies indicate that the proposed approach outperforms the other knowledge transfer methods in efficacy, efficiency, and scalability. We also provide ablation experiments to demonstrate the necessity of the key ingredients.

THE APPLICATION OF DISCUSSION METHOD THROUGH ZOOM MEETING IN THE PANDEMIC TIME IN PKN LEARNING

The pandemic period changes the learning pattern that is usually done offline, now it is done online, so that teachers must make an alternative that is able to carry out learning without the interference of the corona virus. This study aims to improve civics learning outcomes by using the classroom action research method at SMPN 1 Kempas, Kempas District, Indragiri Hilir Regency with 33 students. enough category, the second meeting with a total score of 20 with an average of 71.5 in the enough category. In the second cycle the first meeting with a total score of 21 with an average of 75 in the sufficient category, and the second meeting with a total score of 23 with an average of 82 in the good category. The analysis of student observations in cycle I with an average of 64.3 student activities was categorized as sufficient. The second meeting with an average of 68 student activities was in the moderate category. Whereas at the first meeting of the second cycle with an average of 71.4 student activities could be categorized as sufficient. At the second meeting of cycle II, student activity again increased by an average of 75 and can be categorized as very sufficient. The learning outcomes of this increase can be seen from the first cycle with an average of 71.88 categorized as quite good, and the second cycle with an average percentage of 76.88 categorized good.

Robust Procedures for Estimating and Testing in the Framework of Divergence Measures

The approach for estimating and testing based on divergence measures has become, in the last 30 years, a very popular technique not only in the field of statistics, but also in other areas, such as machine learning, pattern recognition, etc [...]