Knowledge base completion (KBC) aims to predict missing information in a knowledge base. In this paper, we address the out-of-knowledge-base (OOKB) entity problem in KBC: how to answer queries concerning test entities not observed at training time. Existing embedding-based KBC models assume that all test entities are available at training time, making it unclear how to obtain embeddings for new entities without costly retraining. To solve the OOKB entity problem without retraining, we use graph neural networks (Graph-NNs) to compute the embeddings of OOKB entities, exploiting the limited auxiliary knowledge provided at test time. The experimental results show the effectiveness of our proposed model in the OOKB setting. Additionally, in the standard KBC setting in which OOKB entities are not involved, our model achieves state-of-the-art performance on the WordNet dataset.

keywords: information extraction, knowledge base completion, neural network, graph neural network, out-of-knowledgebase entity SummaryKnowledge base completion (KBC) aims to predict missing information in a knowledge base. In this paper, we address the out-of-knowledge-base (OOKB) entity problem in KBC: how to answer queries concerning test entities not observed at training time. Existing embedding-based KBC models assume that all test entities are available at training time, making it unclear how to obtain embeddings for new entities without costly retraining. To solve the OOKB entity problem without retraining, we use graph neural networks (GNNs) to compute the embeddings of OOKB entities, exploiting the limited auxiliary knowledge provided at test time. The experimental results show the effectiveness of our proposed model in the OOKB setting. Additionally, in the standard KBC setting in which OOKB entities are not involved, our model achieves state-of-the-art performance on the WordNet dataset.

No abstract

We develop a general framework for statistical inference with the Wasserstein distance. Recently, the Wasserstein distance has attracted much attention and been applied to various machine learning tasks due to its celebrated properties. Despite the importance, hypothesis tests and confidence analysis with the Wasserstein distance have not been available in a general setting, since a limit distribution of empirical distribution with Wasserstein distance has been unavailable without strong restrictions. In this study, we develop a novel non-asymptotic Gaussian approximation for the empirical Wasserstein distance, which can avoid the problem of unavailable limit distribution. By the approximation method, we develop a hypothesis test and confidence analysis for the empirical Wasserstein distance. We also provide a theoretical guarantee and an efficient algorithm for the proposed approximation. Our experiments validate its performance numerically.

We develop a general framework for statistical inference with the 1-Wasserstein distance. Recently, the Wasserstein distance has attracted considerable attention and has been widely applied to various machine learning tasks because of its excellent properties. However, hypothesis tests and a confidence analysis for it have not been established in a general multivariate setting. This is because the limit distribution of the empirical distribution with the Wasserstein distance is unavailable without strong restriction. To address this problem, in this study, we develop a novel nonasymptotic gaussian approximation for the empirical 1-Wasserstein distance. Using the approximation method, we develop a hypothesis test and confidence analysis for the empirical 1-Wasserstein distance. We also provide a theoretical guarantee and an efficient algorithm for the proposed approximation. Our experiments validate its performance numerically.