Ontologists build formal models to understand the structure of reality. The fun starts –and I had a lot of it back in the PhD days (!)– when Formal Ontology is applied to understand the structure of what we indisputably use to understand reality itself: the mind. Philosophers have spent lifetimes hovering over this conundrum but I stopped more than a decade ago. Fast-forwarding to the present, I have been busy with a not-so-distant, yet more mundane, problem: building ontologies for AI.

My work focuses on engineering ontologies that can be integrated with the artificial minds' “substrata”, i.e. deep and shallow neural networks, and with the processes these bring about, all of which pretty much boil down to pattern recognition. In this keynote I will describe how ontologies can be effectively used in data-driven AI frameworks: I will argue that, in order to progress towards Explainable AI, it is necessary to design hybrid systems that integrate human-accessible machine representations with neural machines. Rather than concocting a philosophical theory, I will build my argument by illustrating core results from some of the projects I have been involved in at Carnegie Mellon first and, more recently, at Bosch.