For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the…
For privacy protection, federated learning based on data separation allows machine learning models to be trained on remote devices or in isolated data devices. However, due to the limited resources such as bandwidth and power of local devices, communication in federated learning can be much slower than in local computing. This study aims to improve communication efficiency by reducing the number of communication rounds and the size of information transmitted in each round.
This paper allows each user node to perform multiple local trainings, then upload the local model parameters to a central server. The central server updates the global model parameters by weighted averaging the parameter information. Based on this aggregation, user nodes first cluster the parameter information to be uploaded and then replace each value with the mean value of its cluster. Considering the asymmetry of the federated learning framework, adaptively select the optimal number of clusters required to compress the model information.
While maintaining the loss convergence rate similar to that of federated averaging, the test accuracy did not decrease significantly.
By compressing uplink traffic, the work can improve communication efficiency on dynamic networks with limited resources.
This paper traces the incorporation of western educational histories in the development of normal-school curricula during the late Qing Dynasty and the Republic of China…
This paper traces the incorporation of western educational histories in the development of normal-school curricula during the late Qing Dynasty and the Republic of China (1901–1944). It uses publication networks to show how the study of comparative educational history facilitated the international circulation of knowledge in the teaching profession, and how the “uses” of educational history were shaped by larger geopolitical forces.
This paper analyzes the international exchange of texts between normal schools in China and Japan and, subsequently, between normal schools in China and the United States. A database of 107 publications in the field of western educational history that were adopted in China reveals specific patterns of textual citation, cross-reference, and canon-formation in the field of educational historiography.
With conclusions derived from a combination of social network analysis and clustering analysis, this paper identifies three broad stages in China's development of normal-school curricula in comparative educational history: “Japan as Teacher,” “transitional period” and “America as Teacher.”
Statistical analysis can reveal citation and reference patterns but not readers' understanding of the deeper meaning of texts – in this case, textbooks on the subject of western educational history. In addition, the types of publications analyzed in this study are relatively limited, the articles on the history of education in journals have not become the main objects of this study.
This paper uses both quantitative and qualitative methods to uncover the transnational circulation of knowledge in the field of comparative educational history during its formative period in China.