No abstract

No abstract

No abstract

Semantic Image Interpretation (SII) is the task of extracting structured semantic descriptions from images. It is widely agreed that the combined use of visual data and background knowledge is of great importance for SII. Recently, Statistical Relational Learning (SRL) approaches have been developed for reasoning under uncertainty and learning in the presence of data and rich knowledge. Logic Tensor Networks (LTNs) are a SRL framework which integrates neural networks with first-order fuzzy logic to allow (i) efficient learning from noisy data in the presence of logical constraints, and (ii) reasoning with logical formulas describing general properties of the data. In this paper, we develop and apply LTNs to two of the main tasks of SII, namely, the classification of an image's bounding boxes and the detection of the relevant part-of relations between objects. To the best of our knowledge, this is the first successful application of SRL to such SII tasks. The proposed approach is evaluated on a standard image processing benchmark. Experiments show that background knowledge in the form of logical constraints can improve the performance of purely data-driven approaches, including the state-of-theart Fast Region-based Convolutional Neural Networks (Fast R-CNN). Moreover, we show that the use of logical background knowledge adds robustness to the learning system when errors are present in the labels of the training data.

Abstract. Semantic coordination, namely the problem of finding an agreement on the meaning of heterogeneous semantic models, is one of the key issues in the development of the Semantic Web. In this paper, we propose a new algorithm for discovering semantic mappings across hierarchical classifications based on a new approach to semantic coordination. This approach shifts the problem of semantic coordination from the problem of computing linguistic or structural similarities (what most other proposed approaches do) to the problem of deducing relations between sets of logical formulae that represent the meaning of concepts belonging to different models. We show how to apply the approach and the algorithm to an interesting family of semantic models, namely hierarchical classifications, and present the results of preliminary tests on two types of hierarchical classifications, web directories and catalogs. Finally, we argue why this is a significant improvement on previous approaches.