Search

One of the biggest challenges in continual learning domains is the tendency of machine learning models to forget previously learned information over time. While overcoming this issue, the existing approaches often exploit large amounts of additional memory and apply model forgetting mitigation mechanisms which substantially prolong the training process. Therefore, we propose a novel SuperFormer method that alleviates model forgetting, while spending negligible additional memory and time. We tackle the continual learning challenges in a learning scenario, where we learn different tasks in a sequential order. We compare our method against several prominent continual learning methods, i.e., EWC, SI, MAS, GEM, PSP, etc. on a set of text classification tasks. We achieve the best average performance in terms of AUROC and AUPRC (0.7% and 0.9% gain on average, respectively) and the lowest training time among all the methods of comparison. On average, our method reduces the total training time by a factor of 5.4-8.5 in comparison to similarly performing methods. In terms of the additional memory, our method is on par with the most memory-efficient approaches. [ABSTRACT FROM AUTHOR]

Air pollutant levels depend on emissions but can also be affected by the meteorological situation. We examined air pollutant trends (PM 10 , NO 2 , O 3 and SO 2 ) in Slovenia, where in the past the main issue were SO 2 levels. Now, the population is still exposed to PM 10 and ozone levels that are above the recommended levels. Our goal was to assess if the levels of air pollutants were decreasing from 2002 to 2017 due to emission ceilings or were more influenced by changes in the meteorological situation. We modelled the relationship between levels, meteorological parameters, and seasonality and then used the models with the best estimated generalisation to adjust levels for meteorology. Models showed a significant relationship between meteorological parameters and PM 10 , NO 2 , and O 3 levels, but not SO 2 . We analysed trends of raw and adjusted levels and compared them. Trends of PM 10 and SO 2 were decreasing at all locations for raw and adjusted data. The largest decrease was observed in SO 2 levels where the largest decrease in emissions occurred. Trends of NO 2 were also significant and negative at most locations. Levels of O 3 did not exhibit a significant trend at most locations. Results show that changes in the meteorological situation affected PM 10 levels the most, especially where the entire period (2002–2017) could be observed. There is strong empirical evidence that changes in meteorological parameters contributed to the decrease in PM 10 levels while the decrease in NO 2 and SO 2 levels can be attributed to emission ceilings. [ABSTRACT FROM AUTHOR]

Nursing clinical learning environments are particularly important for the achievement of good practice in clinical training of student nurses, and thus, for the nursing competence development. Hence, it is important to have an instrument consisting of reliable and valid criteria for assessing the clinical learning environment, applicable in different contexts, and translated in the respondents mother tongue. The goal of the present research was to test the reliability and validity of the Slovenian version of the “Clinical Learning Environment, Supervision and Nurse Teacher evaluation scale”, and to compare it with the Croatian version. The data was collected between 10 March and 10 June 2015 at four Slovenian institutions, where nursing BSc study programmes are performed. The final sample consisted of 232 students (response rate 68.8%): 81.9% were females and 18.1% males, average age was 23. The translated instrument in Slovenian language resulted as reliable and valid, it reflects the expected five factors of the original version despite some minor problems in the factor structure and in test-retest. The most important difference between the Slovenian and Croatian version is in the factor structure regarding the implementation of roles in clinical learning environment. [ABSTRACT FROM AUTHOR]

Polyunsaturated fatty acids (PUFAs) are structural components of membrane phospholipids and precursors of oxygenated lipid mediators with diverse functions, including the control of cell growth, inflammation and tumourigenesis. However, the molecular pathways that control the availability of PUFAs for lipid mediator production are not well understood. Here, we investigated the crosstalk of three pathways in the provision of PUFAs for lipid mediator production: (i) secreted group X phospholipase A 2 (GX sPLA 2) and (ii) cytosolic group IVA PLA 2 (cPLA 2 α), both mobilizing PUFAs from membrane phospholipids, and (iii) adipose triglyceride lipase (ATGL), which mediates the degradation of triacylglycerols (TAGs) stored in cytosolic lipid droplets (LDs). We combined lipidomic and functional analyses in cancer cell line models to dissect the trafficking of PUFAs between membrane phospholipids and LDs and determine the role of these pathways in lipid mediator production, cancer cell proliferation and tumour growth in vivo. We demonstrate that lipid mediator production strongly depends on TAG turnover. GX sPLA 2 directs ω-3 and ω-6 PUFAs from membrane phospholipids into TAG stores, whereas ATGL is required for their entry into lipid mediator biosynthetic pathways. ATGL controls the release of PUFAs from LD stores and their conversion into cyclooxygenase- and lipoxygenase-derived lipid mediators under conditions of nutrient sufficiency and during serum starvation. In starving cells, ATGL also promotes the incorporation of LD-derived PUFAs into phospholipids, representing substrates for cPLA 2 α. Furthermore, we demonstrate that the built-up of TAG stores by acyl-CoA:diacylglycerol acyltransferase 1 (DGAT1) is required for the production of mitogenic lipid signals that promote cancer cell proliferation and tumour growth. This study shifts the paradigm of PLA 2 -driven lipid mediator signalling and identifies LDs as central lipid mediator production hubs. Targeting DGAT1-mediated LD biogenesis is a promising strategy to restrict lipid mediator production and tumour growth. [Display omitted] • Phospholipase A 2 -induced lipid mediator production depends on lipid droplet turnover. • Group X sPLA 2 -released polyunsaturated fatty acids are incorporated into triglycerides. • Triglyceride lipolysis by ATGL drives lipid mediator production. • ATGL also delivers polyunsaturated fatty acids into phospholipids to be targeted by cPLA 2 α. • Inhibition of DGAT1 reduces cancer cell proliferation and tumour growth in vivo. [ABSTRACT FROM AUTHOR]

Summary: Background: Both academics and practitioners agree that critical thinking skills are necessary to provide safe and comprehensive nursing care. In order to promote the development of critical thinking, nurse educators need to keep the teaching/learning process captivating and interesting using active learning environments. These can be implemented by using modern information and communication technologies that are simple, fun, and time and cost effective. Objectives: The goal of our study was to design and test an approach, which allows individual and fast acquisition of critical thinking skills with the use of information and communication technology. Design: A combination of qualitative and quantitative research design was implemented. The study consisted of a quasi-experiment (phases 1–3): (1) pre-test discussion board, (2) use of e-contents based on the presented approach, and (3) post-test discussion board. The participants' opinion about the presented approach was identified in phase 4. Settings: The study was performed in May 2012 during the course “Ethics and Philosophy in Nursing” at the Faculty of Health Sciences, University of Primorska, Slovenia. Participants: Forty first-year undergraduate nursing students. Methods: Qualitative analysis of the discussion boards (phases 1, 3) and an anonymous survey with open- and closed-ended questions (phase 4). Results: Qualitative analysis of the discussion boards showed a significant (p<0.001) improvement in the percentage of posts (12.2%) for which the opinions and conclusions of the participants were justified with valid arguments. The survey results indicated that participants perceived the e-contents based on the presented approach as useful, and that they improved their critical thinking skills. Conclusions: Repeated confirmation of the validity of the presented approach through methodological triangulation represents a strong indication that the presented approach is a valuable tool to develop nursing students' critical thinking skills. [Copyright &y& Elsevier]

Summary: Although e-learning is an innovation that is worth making generally available, it is not always accepted by nursing students. Many researchers state that voluntariness is closely related to the individual level of adoption of innovations. Hence, we hypothesized that voluntariness moderates the effect of perceived attributes of innovations (e.g. relative advantage, compatibility, complexity, trialability, and observability), which determines the acceptance of e-learning. To test the hypothesis a survey involving two groups of nursing students was carried out. For the first group the usage of e-learning was mandatory, for the second group it was optional. The results confirm our hypothesis. Institutions, interested in e-learning initiatives, should consider the effect of voluntariness when implementing e-learning. This paper provides a useful reference that can help e-learning providers to develop guidelines that can improve the acceptance of e-learning. [Copyright &y& Elsevier]