We are happy to announce that our paper “Causality Prediction with Neural-Symbolic Systems: A Case Study in Smart Grids” has been accepted as a workshop paper to be presented at the 17th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy2023), taking place July 3-5 in Certosa di Pontigamo, Siena, Italy. 

The goal of the paper is to improve the explainability of complex systems, such as smart grids through neural-symbolic AI. Explaining events in these systems is crucial for both system operators and users. Traditional symbolic AI systems, which have been used to address this issue in the past, have a limited scope due to their rule-based approach. To address this, our paper proposes a neural-symbolic architecture, which combines symbolic approaches with sub-symbolic components. Knowledge Graph Embeddings are used to represent causal relations between events and to predict additional causal relations, which may have been missed by the purely symbolic system. The experimental results show that this approach has the potential to improve the performance of causality detection in smart grids and other complex systems.