Distantly Supervised Relation Extraction in Federated Settings
This work addresses data barriers and privacy concerns in relation extraction for distributed platforms, offering a practical solution but is incremental as it adapts existing denoising methods to a federated context.
The paper tackles the problem of distantly supervised relation extraction in federated settings, where label noise is exacerbated by data scattering across platforms, and proposes a federated denoising framework that uses cross-platform collaboration to select reliable instances, achieving effective results as demonstrated on the New York Times and miRNA gene regulation datasets.
This paper investigates distantly supervised relation extraction in federated settings. Previous studies focus on distant supervision under the assumption of centralized training, which requires collecting texts from different platforms and storing them on one machine. However, centralized training is challenged by two issues, namely, data barriers and privacy protection, which make it almost impossible or cost-prohibitive to centralize data from multiple platforms. Therefore, it is worthy to investigate distant supervision in the federated learning paradigm, which decouples the model training from the need for direct access to the raw data. Overcoming label noise of distant supervision, however, becomes more difficult in federated settings, since the sentences containing the same entity pair may scatter around different platforms. In this paper, we propose a federated denoising framework to suppress label noise in federated settings. The core of this framework is a multiple instance learning based denoising method that is able to select reliable instances via cross-platform collaboration. Various experimental results on New York Times dataset and miRNA gene regulation relation dataset demonstrate the effectiveness of the proposed method.