Search

Vision Transformers (ViTs) have proved to be a more powerful substitute for Convolutional Neural Networks (CNNs) in various computer vision tasks, using the self-attention approach to gain remarkable results and observations. However, the adversarial robustness of ViTs against adversarial attack methods raises critical questions, and the issues of using these models in security-related applications remain under discussion. This paper presents a novel and systematic approach to evaluate and compare the adversarial robustness of ViTs with CNNs, explicitly concentrating on the image classification problem. We have performed extensive experiments using state-of-the-art adversarial example attacks, such as the Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), and DeepFool Attack (DFA). The findings of this research study represent that CNNs are more robust against more straightforward attacks such as FGSM. Still, ViTs show excellent resistance against more dangerous attacks like PGD and DFA attack methods. This work provides useful outcomes revealing the advantages and limitations of CNNs and ViTs, which are helpful for further study and applications regarding safer and more effective use of deep learning models of CNNs and ViTs.

Purpose: The present study investigates the influence of demographic factors on the demand for Technical and Vocational Education and Training (TVET) in Pakistan. The government of Pakistan has implemented various skill enhancement programs to harness the demographic dividend. However, only a small portion of the workforce receives any form of TVET, contributing to a shortage of skilled workers in the country. Many industries, particularly in manufacturing and mining, face deficits in the skilled labour. Consequently, this study aims to examine the role of demographic factors in shaping the demand for TVET within the Pakistani context. Methods: For the TVET demand's estimation, this study employed the Pakistan Social and Living Standard Measurement (PSLM) dataset of 2018–19 by using binary logistic regression analysis (BLRA). The demographic variables include the household's income, household head's education, household size, male proportion of the target age group, household head's age, and region of the household. Findings: The findings indicate that households in the higher income category do not demand TVET. Moreover, if the head of the household is highly educated, then the household is less likely to participate in TEVT. So, the higher the socio-economic status, the lower the probability of demand for TVET from the better-off students. Further, this study also indicates that boys are more likely to participate in TVET-related degrees, while females are less likely to participate in TVET due to the non-availability of institutes and hostel facilities, poor transportation, the limited number of trades available for females, and security issues. Conclusion: The findings provide insightful evidence to support the idea that the higher the socio-economic status of households, the lower the probability of demand for a TVET degree or diploma. Similarly, children of parents with university education are less likely to pursue TVET-related degrees. The reason is likely attributed to the perception that TVET-associated degrees and diplomas are considered inferior due to their lower standing and prestige as compared to general or professional degrees. This study suggests that the attractiveness of TVET can be enhanced by improving the quality of TVET, improving labour market outcomes, and creating a pathway to general education. Overall, this study not only contributes to empirical analyses of socio-economic determinants in TVET demand but also suggests that its findings can be applied not only to South Asian countries but also to other comparable nations with similar cultural ties and affinities.

Potato is one of the most important vegetable crops worldwide. Its growth, development and ultimately yield is hindered by drought stress condition. Breeding and selection of deep-rooted and drought-tolerant potato varieties has become a prime approach for improving the yield and quality of potato (Solanum tuberosum L.) in arid and semiarid areas. A comprehensive understanding of root development-related genes has enabled scientists to formulate strategies to incorporate them into breeding to improve complex agronomic traits and provide opportunities for the development of stress tolerant germplasm. Root response to drought stress is an intricate process regulated through complex transcriptional regulatory network. To understand the rooting depth and molecular mechanism, regulating root response to drought stress in potato, transcriptome dynamics of roots at different stages of drought stress were analyzed in deep (C119) and shallow-rooted (C16) cultivars. Stage-specific expression was observed for a significant proportion of genes in each cultivar and it was inferred that as compared to C16 (shallow-rooted), approximately half of the genes were differentially expressed in deep-rooted cultivar (C119). In C16 and C119, 11 and 14 coexpressed gene modules, respectively, were significantly associated with physiological traits under drought stress. In a comparative analysis, some modules were different between the two cultivars and were associated with differential response to specific drought stress stage. Transcriptional regulatory networks were constructed, and key components determining rooting depth were identified. Through the results, we found that rooting depth (shallow vs deep) was largely determined by plant-type, cell wall organization or biogenesis, hemicellulose metabolic process, and polysaccharide metabolic process. In addition, candidate genes responding to drought stress were identified in deep (C119) and shallow (C16) rooted potato varieties. The results of this study will be a valuable source for further investigations on the role of candidate gene(s) that affect rooting depth and drought tolerance mechanisms in potato.