Knowledge graph (KG) embedding is to embed components of a KG including entities and relations into continuous vector spaces, so as to simplify the manipulation while preserving the inherent ...structure of the KG. It can benefit a variety of downstream tasks such as KG completion and relation extraction, and hence has quickly gained massive attention. In this article, we provide a systematic review of existing techniques, including not only the state-of-the-arts but also those with latest trends. Particularly, we make the review based on the type of information used in the embedding task. Techniques that conduct embedding using only facts observed in the KG are first introduced. We describe the overall framework, specific model design, typical training procedures, as well as pros and cons of such techniques. After that, we discuss techniques that further incorporate additional information besides facts. We focus specifically on the use of entity types, relation paths, textual descriptions, and logical rules. Finally, we briefly introduce how KG embedding can be applied to and benefit a wide variety of downstream tasks such as KG completion, relation extraction, question answering, and so forth.
Relational machine learning studies methods for the statistical analysis of relational, or graph-structured, data. In this paper, we provide a review of how such statistical models can be "trained" ...on large knowledge graphs, and then used to predict new facts about the world (which is equivalent to predicting new edges in the graph). In particular, we discuss two fundamentally different kinds of statistical relational models, both of which can scale to massive data sets. The first is based on latent feature models such as tensor factorization and multiway neural networks. The second is based on mining observable patterns in the graph. We also show how to combine these latent and observable models to get improved modeling power at decreased computational cost. Finally, we discuss how such statistical models of graphs can be combined with text-based information extraction methods for automatically constructing knowledge graphs from the Web. To this end, we also discuss Google's knowledge vault project as an example of such combination.
Generating fit-for-purpose CAE models from complex CAD assemblies is time consuming and error-prone. Tedious tasks include identifying and isolating the components of interest, removing duplicate ...components, and correcting inconsistent component interfaces. In this paper a new approach to help engineers identify similar features and analyse the consistency of CAD assembly models is proposed. The method utilises a tensor factorisation technique developed for relational machine learning and applies it to B-Rep topological and geometrical relations. The model considers globally all the input relations to identify which entities in the assembly are similar (within a user-defined threshold) to a selected input entity. It is shown that a hierarchical clustering method can group entities, based on the similarities of their attributes and relationships with adjacent components. It is shown how some unsuspected CAD modelling errors show up as features which should be similar, but which are not. It is demonstrated how the technique can be used to support the, currently highly manual, task of decomposing a volume representing an internal fluid network into sub-volumes and features of significance.
•Application of relational learning tensor factorisation to B-Rep models.•New approach to analyse the consistency of large CAD assemblies.•Usage scenario for CAD/CAE integration: Finding similar features in an assembly.•Hierarchical clustering of similar CAD parts in an assembly.•Industrial use case: Finding airflow features between parts in a complex assembly.
Knowledge graphs are multi-relational data that contain massive entities and relations. As an effective graph representation technique based on deep learning, graph neural network has reported ...outstanding performance for modeling knowledge graphs in recent studies. However, previous graph neural network-based models have not fully considered the heterogeneity of knowledge graphs. Furthermore, the attention mechanism has demonstrated its great potential in many areas. In this paper, a novel heterogeneous graph neural network framework based on a hierarchical attention mechanism is proposed, including entity-level, relation-level, and self-level attentions. Thus, the proposed model can selectively aggregate informative features and weights them adequately. Then the learned embeddings of entities and relations can be utilized for the downstream tasks. Extensive experimental results on various heterogeneous graph tasks demonstrate the superior performance of the proposed model compared to several state-of-the-art methods.
Most knowledge graph completion methods focus on predicting existing relationships in the knowledge graph but cannot predict unseen relationships. To solve this problem, knowledge graph zero-shot ...relational learning (KGZSL) has gotten more and more attention in recent years. The common method of KGZSL is to leverage Generative Adversarial Networks (GANs) to build the connection between existing relation descriptions and knowledge graph domains. However, the traditional KGZSL method ignores the gap between relation text description and relation structured representation. To bridge this gap, we propose a Structure-Enhanced Generative Adversarial Network (SEGAN). SEGAN adopts a structure encoder to introduce knowledge graph structure information into the generator and guide the generator to generate knowledge graph embeddings more accurately. In addition, in the KGZSL task, the representations of entities are closely tied to the existing relationships, which has a negative impact on the prediction of new instances. Therefore, we design a new plug-and-play feature encoder to decouple entities from existing relationships. Experimental results on the knowledge graph zero-shot relational learning dataset demonstrate that our method has better structure representation ability, and the model performance is improved by 36.3% compared with the current optimal model.
A knowledge graph (KG), also known as a knowledge base, is a particular kind of network structure in which the node indicates entity and the edge represent relation. However, with the explosion of ...network volume, the problem of data sparsity that causes large-scale KG systems to calculate and manage difficultly has become more significant. For alleviating the issue, knowledge graph embedding is proposed to embed entities and relations in a KG to a low-, dense and continuous feature space, and endow the yield model with abilities of knowledge inference and fusion. In recent years, many researchers have poured much attention in this approach, and we will systematically introduce the existing state-of-the-art approaches and a variety of applications that benefit from these methods in this paper. In addition, we discuss future prospects for the development of techniques and application trends. Specifically, we first introduce the embedding models that only leverage the information of observed triplets in the KG. We illustrate the overall framework and specific idea and compare the advantages and disadvantages of such approaches. Next, we introduce the advanced models that utilize additional semantic information to improve the performance of the original methods. We divide the additional information into two categories, including textual descriptions and relation paths. The extension approaches in each category are described, following the same classification criteria as those defined for the triplet fact-based models. We then describe two experiments for comparing the performance of listed methods and mention some broader domain tasks such as question answering, recommender systems, and so forth. Finally, we collect several hurdles that need to be overcome and provide a few future research directions for knowledge graph embedding.
Markov Logic Networks (MLNs) are discrete generative models in the exponential family. However, specifying these rules requires considerable expertise and can pose a significant challenge. To ...overcome this limitation, Neural MLNs (NMLNs) have been introduced, enabling the specification of potential functions as neural networks. Thanks to the compact representation of their neural potential functions, NMLNs have shown impressive performance in modeling complex domains like molecular data. Despite the superior performance of NMLNs, their theoretical expressiveness is still equivalent to that of MLNs without quantifiers. In this paper, we propose a new class of NMLN, called Quantified NMLN, that extends the expressivity of NMLNs to the quantified setting. Furthermore, we demonstrate how to leverage the neural nature of NMLNs to employ learnable aggregation functions as quantifiers, increasing expressivity even further. We demonstrate the competitiveness of Quantified NMLNs over original NMLNs and state-of-the-art diffusion models in molecule generation experiments.
The extreme version of the Whorfian hypothesis-that the language we learn determines how we view the world-has been soundly rejected by linguists and psychologists alike. However, more moderate ...versions of the idea that language may influence thought have garnered recent empirical support. This article defends 1 such view. I propose that language serves as a cognitive tool kit that allows us to represent and reason in ways that would be impossible without such a symbol system. I present evidence that learning and using relational language can foster relational reasoning-a core capacity of higher order cognition. In essence, language makes one smarter.
Lifted inference with tree axioms van Bremen, Timothy; Kuželka, Ondřej
Artificial intelligence,
November 2023, 2023-11-00, Volume:
324
Journal Article
Peer reviewed
We consider the problem of weighted first-order model counting (WFOMC): given a first-order sentence ϕ and domain size n∈N, determine the weighted sum of models of ϕ over the domain {1,…,n}. Past ...work has shown that any sentence using at most two logical variables admits an algorithm for WFOMC that runs in time polynomial in the given domain size 1,2. The same property was later also shown to hold for C2, the two-variable fragment with counting quantifiers 3. In this paper, we further expand this result to any C2 sentence ϕ with the addition of a tree axiom, stating that some distinguished binary relation in ϕ forms a tree in the graph-theoretic sense.