Thinking on a Grand Scale

Lecturer: Prof. Michael Witbrock (Cycorp Inc., USA).

About the lecturer | Course Summary | Slides | Assignment

About the lecturer: Dr. Michael Witbrock serves as the Vice President for Research at Cycorp, Inc. and as CEO of Cycorp Europe. At Cycorp, he has overall responsibility for corporate research, and is particularly interested in automating the process of knowledge acquisition and elaboration, extending the range of knowledge representation and reasoning to mixed logical and probabilistic representations, and in validating and elaborating knowledge in the context of task performance, particularly in tasks that involve understanding text and communicating with users. Michael received his PhD in Computer Science from Carnegie Mellon in 1996 and a BSc in Psychology from Otago University, in New Zealand, in 1985. Prior to joining Cycorp, he was Principal Scientist at Terra Lycos, working on integrating statistical and knowledge based approaches to understanding web user behavior; a research scientist at Just Systems Pittsburgh Research Center, working on statistical text summarization; and a systems scientist at Carnegie Mellon on the Informedia spoken and video document information retrieval project. He is author of numerous publications in areas ranging across knowledge representation and acquisition, neural networks, parallel computer architecture, multimedia information retrieval, web browser design, genetic design, computational linguistics and speech recognition, and is the holder of four US patents.

Course summary:
There is currently a renaissance in the application of logic-based AI techniques, some of it under the rubric of the "Semantic Web". But another dark age is possible if these techniques cannot be applied at scale, in terms of knowledge capture, breadth of coverage, and scale of reasoning. None of these have yet been demonstrated at the required levels, but some promising approaches exist. In the first part of the course, we will characterise these problems more precisely and then explore how representations from RDF to probabilistic logics affect potential breadth, diving more deeply into the representational capabilities of ResearchCyc. We'll also discuss and use knowledge capture techniques that may make it possible to apply those representations at social content (Web 2.0) scale. In the second part of the course, we'll discuss scaling reasoning, both in terms of inferential power and in terms of inferential bulk, concentrating on the goals and techniques of the FP7 LarKC project, and on using inference with Research Cyc and OpenCyc. The course will be fairly self-contained, but some familiarity with first order logic and web services will be helpful, as will some familiarity with Java (although PHP, C# or Ruby should suffice).


Assignment is here: [PDF]. Due: 24 May (please send pdfs to Michael Witbrock).
Results will be discussed during the second part of the classes.