Knowledge Graphs: Bringing Logical and Statistical Reasoning Together
【Abstract】
Knowledge graphs (KGs) have in recent years gained a large momentum both in academic research and in real-world applications. They have become a bridge between databases, artificial intelligence (AI), data science, and many other areas. One particularly interesting aspect is that they have become a link between the more logical reasoning-oriented area of databases and the more statistical reasoning-oriented areas of machine learning.
In this talk, I will give an overview of both aspects: logical and statistical.
I will start by giving an overview of logical languages for Knowledge Graphs, in particular the Vadalog language developed at Oxford, TU Wien and partners.
Then I will continue on statistical aspects of Knowledge Graphs, chiefly Knowledge Graph Embeddings, and, briefly, Graph Neural Networks.
I will show both the theory of it, but also how to see it "in action" based on multiple real-world applications, such as some of the following: corporate governance, media intelligence, supply chains, collateral eligibility,hostile takeovers, smart anonymization, and anti-money laundering.
【Bio】
Emanuel Sallinger is Professor at TU Wien and Lecturer for Knowledge Graphs and Databases at Oxford University. He is heading the Knowledge Graph Lab at TU Wien and is lead of the SIG Knowledge Graphs at the Center for AI and ML (CAIML). His main research focus is on Knowledge Graphs, including all theoretical and practical aspects. In particular, he is interested in reasoning in such systems, including all of the AI methodologies for that (knowledge-based/logic-based reasoning and statistics/ML-based reasoning). Within such systems, his interest is in achieving scalable solutions, making sure that theory translates into practice.
