Your conditions: 西安电子科技大学
  • 从生活空间到育人空间——Z世代大学生群体特征及其对高校现代书院人才培养的启示

    Subjects: Other Disciplines >> Synthetic discipline submitted time 2024-03-05 Cooperative journals: 《第八届海峡两岸暨港澳地区高校现代书院制教育论坛 》

    Abstract:文章主要论述了Z世代群体的定义及其五个特征:Z世代大学生的“网络原住民”特性、网络爱表达特性,但不人云亦云,有自己的独立见解、兴趣爱好非常广泛、具有强烈的家国情怀和爱国意识,全球视野,具有平视世界的信心,同时追求较高的认同感、有较强的沉浸式体验感。高校现代书院建设需要实施从“生活空间”向“文化空间”最终向“育人空间”的转向,融合“寓教育与生活,生活中育人”的理念。要立足于对学生“可迁移能力”和“未来自我发展能力”的培养,从而为学生的未来做准备。

  • A Novel Two-Party Comparison Protocol Against Untrusted Parties

    Subjects: Computer Science >> Information Security submitted time 2023-04-08

    Abstract: Secure two-party comparison is widely used to build various secure computing protocols (e.g., secure training, secure inference). In existing secure two-party comparison protocols, there is always one party that obtains a comparison result first, and then the party notifies the comparison result to the other one, thus, they are difficult to prevent one party that obtains the comparison result first from tampering with the comparison result. To this end, this paper first proposes a new paradigm for secure two-party comparison against untrusted parties. Then, a secure two-party comparison protocol (TOMS) satisfying the new paradigm is designed based on the threshold Paillier cryptosystem. Each party in TOMS obtains the same comparison result without revealing their own data. Moreover, TOMS prevents any party from tampering with the comparison results. Strictly theoretical analyses demonstrate the security and correctness of TOMS. Finally, the experimental results show that TOMS outperforms the existing secure two-party comparison methods in terms of computational efficiency and functionality, and is 50 times faster than previous methods.

  • Strengthen Fundamental Role of Data Element Governance in National Governance Modernization

    Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-28 Cooperative journals: 《中国科学院院刊》

    Abstract: Data element governance is a key factor to promote the modernization of national governance in the digital era. By strengthening the deep integration of data factors and national governance, a new model of data-driven national governance can be formed, and the national governance can be made more scientific, refined, intelligent, and efficient. The US and European countries have continuously strengthened the top-level system design, technological innovation application, collaborative governance mechanism, and global governance cooperation of data element governance, which has effectively improved the level of data element governance and provided experience for China. Nevertheless, due to the virtuality of data elements, more subjects involved, greater risk of leakage, and higher requirements of technical support level, the current data element governance is still faced with challenges such as imperfect collaborative governance mechanism, incomplete ethical governance mechanism, weak talent and technology foundation, etc. It is thus challenging to give full play to the role of data factors to meet the needs of the modernization of national governance. In the new era, it is urgent to establish and improve a data element governance system suitable for current major needs and application scenarios of national governance, by establishing a governance concept that takes into account the development of efficiency, fairness, and security in all directions, building a governance model featuring the collaboration and co-governance of multiple subjects of the government, the market, and society, strengthening a governance idea driven by the synergy of science, technology, and institutions, and improving the basic guarantee of talents, infrastructure, and public services, to promote the governance level of data factors and provide strong support for the modernization of national governance system and capacity.

  • Digital Technology Enables Construction of National Governance Modernization

    Subjects: Other Disciplines >> Synthetic discipline submitted time 2023-03-28 Cooperative journals: 《中国科学院院刊》

    Abstract: As digital technologies continue to be integrated into the whole process of economic and social development, promoting the modernization of digital technology-enabled national governance systems and capabilities has become an important way to seize the strategic initiative in the future world competitive landscape, and has attracted the attention of countries around the world. The rapid development of digital technologies such as big data collection, storage, processing, and analysis is constantly optimizing the organizational system structure of national governance, upgrading and perfecting the quality and methods of national governance personnel, and accelerating the process of making national governance efficient, scientific, intelligent and refined. At present, the national governance of China has made remarkable achievements but still faces many problems and challenges. The development of digital technologies for national governance needs scene expansion and talent supply, and the negative effects of digital technologies put new requirements on national governance. Based on the current development needs of the modernization of national governance supported by digital technology, this study puts forward suggestions to promote the modernization of national governance system and governance capacity endowed by digital technology, and quicken the pace to reach the overall goal of realizing the modernization of national governance.

  • Geometric Prior Guided Feature Representation Learning for Long-Tailed Classification

    Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2023-02-16

    Abstract: Real-world data are long-tailed, the lack of tail samples leads to a significant limitation in the generalization ability of the model. Although numerous approaches of class re-balancing perform well for moderate class imbalance problems, additional knowledge needs to be introduced to help the tail class recover the underlying true distribution when the observed distribution from a few tail samples does not represent its true distribution properly, thus allowing the model to learn valuable information outside the observed domain. In this work, we propose to leverage the geometric information of the feature distribution of the well-represented head class to guide the model to learn the underlying distribution of the tail class. Specifically, we first systematically define the geometry of the feature distribution and the similarity measures between the geometries, and discover four phenomena regarding the relationship between the geometries of different feature distributions. Then, based on four phenomena, feature uncertainty representation is proposed to perturb the tail features by utilizing the geometry of the head class feature distribution. It aims to make the perturbed features cover the underlying distribution of the tail class as much as possible, thus improving the model’s generalization performance in the test domain. Finally, we design a three-stage training scheme enabling feature uncertainty modeling to be successfully applied. Experiments on CIFAR-10/100-LT, ImageNet-LT, and iNaturalist2018 show that our proposed approach outperforms other similar methods on most metrics. In addition, the experimental phenomena we discovered are able to provide new perspectives and theoretical foundations for subsequent studies. The code will be available at https://github.com/mayanbiao1234/Geometric-Prior

  • 基于SDWN的负载感知终端多点关联方案研究

    Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-04-17 Cooperative journals: 《计算机应用研究》

    Abstract: Software Defined Wireless Network (SDWN) is a wireless network architecture that separates control plane and forward plane. It can get global topology quickly and its programmable combined with resource virtualization technology can control network access dynamically. In order to solve the problem of unbalanced load in existing WLAN and handover problem caused by the hard association between mobile terminal and AP, this paper proposes a terminal multi-point association scheme based on load-aware in SDWN. This scheme improves the seamless handoff algorithm between AP based on traffic load perception. When sensing AP load, the terminal can connect multiple virtual APs to achieve dynamic diversion. Finally, this scheme is verified on an experimental platform. Experiments show that this scheme changes the relationship between the terminal and the AP one by one compared to seamlessly switching to another AP, which effectively improves the terminal network throughput.