Search results

While many scholars find that security cooperation increases the foreign aid that allies receive from a major power, other studies show that an alliance also has negative effect and its terms can change over time. This article argues that a donor's security environment impacts its participation in security cooperation and, subsequently, foreign aid allocation to allies. When the security environment is competitive, a donor will concentrate its resources on strengthening its ties with allies. In contrast, when the environment is favorable to a donor, it can use security cooperation as leverage to obtain policy concessions from allies, thus reducing its own foreign aid expenditures. This paper analyzes the aid allocation of three major power donors as well as 72 formal defense pacts and 70 bilateral defense cooperation agreements, and it finds that a favorable security environment allows a donor to reduce its economic contribution to its allies by 19 percent.

While many scholars find that security cooperation increases the foreign aid that allies receive from a major power, other studies show that an alliance also has negative effect and its terms can change over time. This article argues that a donor's security environment impacts its participation in security cooperation and, subsequently, foreign aid allocation to allies. When the security environment is competitive, a donor will concentrate its resources on strengthening its ties with allies. In contrast, when the environment is favorable to a donor, it can use security cooperation as leverage to obtain policy concessions from allies, thus reducing its own foreign aid expenditures. This paper analyzes the aid allocation of three major power donors as well as 72 formal defense pacts and 70 bilateral defense cooperation agreements, and it finds that a favorable security environment allows a donor to reduce its economic contribution to its allies by 19 percent.

Aviation hubs are pivotal intermediate nodes for connecting people around the world. However, there has been limited investigation on the competitive hub landscape in the China–Oceania market. Employing an updated Connectivity Utility Model, we examine the hub connections provided by hubs in mainland China, Hong Kong and Macau, other Asian countries and Oceania from quantity and quality dimensions between 1997 and 2019. The results indicate that the solid growth of mainland China and Oceania hubs suppress the development of Hong Kong and other Asian hubs. The major competitions of the mainland China and Oceanian hubs have shifted from competing with hubs outside their respective regions to competing with hubs within their regions, while the primary competitors of other Asian hubs have changed from competing with each other to competing with hubs outside their region. In the monopoly market, the Oceanian hubs have always enjoyed the greatest monopoly superiority. In the overlapping market, the dominance of Hong Kong and other Asian hubs in both the quantity and quality dimensions has been challenged by mainland China hubs. In a semiregulated operating environment shaped by China's centralized government, the coordination between airlines, airports and governments is crucial for future hub construction.

Deploying deep learning services for time-sensitive and resource-constrained settings such as IoT using edge computing systems is a challenging task that requires dynamic adjustment of inference time. Multi-exit architectures allow deep neural networks to terminate their execution early in order to adhere to tight deadlines at the cost of accuracy. To mitigate this cost, in this paper we introduce a novel method called Multi-Exit Curriculum Learning that utilizes curriculum learning, a training strategy for neural networks that imitates human learning by sorting the training samples based on their difficulty and gradually introducing them to the network. Experiments on CIFAR-10 and CIFAR-100 datasets and various configurations of multi-exit architectures show that our method consistently improves the accuracy of early exits compared to the standard training approach. ; This work was partly funded by the European Union's Horizon 2020 research and innovation programme under grant agreement No 957337, and by the Danish Council for Independent Research under Grant No. 9131-00119B. This publication reflects the authors views only. The European Commission and the Danish Council for Independent Research are not responsible for any use that may be made of the information it contains.

Deep neural networks can be converted to multi- exit architectures by inserting early exit branches after some of their intermediate layers. This allows their inference process to become dynamic, which is useful for time critical IoT applica- tions with stringent latency requirements, but with time-variant communication and computation resources. In particular, in edge computing systems and IoT networks where the exact computa- tion time budget is variable and not known beforehand. Vision Transformer is a recently proposed architecture which has since found many applications across various domains of computer vision. In this work, we propose seven different architectures for early exit branches that can be used for dynamic inference in Vision Transformer backbones. Through extensive experiments involving both classification and regression problems, we show that each one of our proposed architectures could prove useful in the trade-off between accuracy and speed. ; This work was funded by the European Union's Horizon 2020 research and innovation programme under grant agreement No 957337, and by the Danish Council for Independent Research under Grant No. 9131-00119B. This publication reflects the authors views only. The European Commission and the Danish Council for Independent Research are not responsible for any use that may be made of the information it contains.

Student organization is considered an important approach for the sustainable development of ideological and political education, and it helps construct learning style and campus culture. This research studies college students&rsquo ; motivations for their involvement and exit behaviors. Using a binary choice model, we analyze critical factors affecting the involvement and exit behaviors of college students based on social cognition theory, Maslow&rsquo ; s hierarchy of needs, and expectancy theory. On the one hand, our results show active cognition is most influential for the involvement behavior, followed by personal interest. Meanwhile, active cognition is positively affected by mother&rsquo ; s higher education level and senior students&rsquo ; influence. Respondents who agree that joining organizations brings closer relations with teachers and other students as well as promoting heterosexual communication are more likely to join. On the other hand, the most influential factor for the exit behavior is unclear promotion mechanisms. Low-quality activities and wasting time with organizations cause students to quit. Moreover, lack of training opportunities and professional guidance dampens members&rsquo ; enthusiasm. Complicated interpersonal relations, bureaucratic style, entering a higher grade and changing psychology cause some members to quit and further affect organizational learning. We provide additional insights on management strategies towards organization sustainability, such as improving leader selection and building harmonious interpersonal relations.