Yaobin He’s scientific contributions

Federated learning (FL) and blockchains exhibit significant commonality, complementarity, and alignment in various aspects, such as application domains, architectural features, and privacy protection mechanisms. In recent years, there have been notable advancements in combining these two technologies, particularly in data privacy protection, data sharing incentives, and computational performance. Although there are some surveys on blockchain-based federated learning (BFL), these surveys predominantly focus on the BFL framework and its classifications, yet lack in-depth analyses of the pivotal issues addressed by BFL. This work aims to assist researchers in understanding the latest research achievements and development directions in the integration of FL with blockchains. Firstly, we introduced the relevant research in FL and blockchain technology and highlighted the existing shortcomings of FL. Next, we conducted a comparative analysis of existing BFL frameworks, delving into the significant problems in the realm of FL that the combination of blockchain and FL addresses. Finally, we summarized the application prospects of BFL technology in various domains such as the Internet of Things, Industrial Internet of Things, Internet of Vehicles, and healthcare services, as well as the challenges that need to be addressed and future research directions.

Text classification is a well-established task in NLP, but it has two major limitations. Firstly, text classification is heavily reliant on domain-specific knowledge, meaning that a classifier that is trained on a given corpus may not perform well when presented with text from another domain. Secondly, text classification models require substantial amounts of annotated data for training, and in certain domains, there may be an insufficient quantity of labeled data available. Consequently, it is essential to explore methods for efficiently utilizing text data from various domains to improve the performance of models across a range of domains. One approach for achieving this is through the use of multi-domain text classification models that leverage adversarial training to extract domain-shared features among all domains as well as the specific features of each domain. After observing the varying distinctness of domain-specific features, our paper introduces a curriculum learning approach using a ranking system based on keyword weight to enhance the effectiveness of multi-domain text classification models. The experimental data from Amazon reviews and FDU-MTL datasets show that our method significantly improves the efficacy of multi-domain text classification models adopting adversarial learning and reaching state-of-the-art outcomes on these two datasets.