I am interested in artificial intelligence and its application to real-world problems. Over the years, I have worked in many areas of AI, including knowledge representation, planning, image processing, machine learning, massively parallel approaches to search and inference, and the development of improved learning algorithms for artificial neural networks.
My current project is Scone, an open-source knowledge representation system and inference engine that can be used as a component in a variety of knowledge-based systems. Scone puts particular emphasis on efficiency, scalability (up to millions of entities and statements) and ease of use. One goal of our research is to support natural-language understanding, all the way from text or speech to a language-independent representation that we can reason about. This requires extensive use of background knowledge not included in the text itself. We are also working to extend Scone's capabilities for "episodic" representation and reasoning, by which we mean actions, events, sequences of actions, plans, goals and explanations.
I have also worked on programming languages and software-development environments that support an incremental, evolutionary software-development style. This work has been done in Common Lisp, Dylan and Java.