Staff Page - John Maraist

Dr. Maraist is a SIFT Research Scientist specializing in Artificial Intelligence and in particular planning and scheduling. His areas of expertise also include formal techniques for programming languages and software engineering.

Currently Dr. Maraist is part of SIFT's efforts on the DARPA program "National Cyber Range" for developing a standard national testbed for computer and network security. Dr. Maraist has previously worked on three other large DARPA programs: Integrated Learning (2006-2010) for recognizing and learning user intent from single-shot observation of actions; Deep Green (2008-2009) for augmenting in-progress battlefield tactical analysis with scenario simulation, and Self-Reconfiguring Systems (SRS, 2006-2007) for defense against insider threats on computer networks. He has also participated in a number of SBIR programs, including the Phase II DARPA SBIR ELADIS (Extremely Low Attention Demand Information Systems, 2009-2010).

Dr. Maraist is the primary developer of the NST unit testing framework for Common Lisp. NST was first developed for the Integrated Learning program, has since been used successfully on other SIFT programs including SRS and SMITE, and is freely available under an open-source license.

Prior to joining SIFT, Dr. Maraist was a consultant at Model Software Corporation, where he developed a generalized logic engine for rapid deployment of specialized scheduling solutions. He has also worked on the faculties of DePaul University and the University of South Australia, pursuing research in programming language semantics.

Selected publications

• Two short papers on Shopper, an interpreter for the planning language LTML, both written with Robert Goldman:
  • "Shopper: a system for executing and simulating expressive plans", Proceedings of the International Conference on Automated Planning and Scheduling (ICAPS'10), April 2010. This paper discusses Shopper's relevance to planning, especially in the context of web service composition.
  • "Shopper: interpreter for a high-level web services language", Proceedings of the International Lisp Conference, March 2009. This paper focuses on the software engineering of the system, in particular the advantages we enjoyed using Lisp as the implementation language.
• "LTML — A Language for Representing Semantic Web Service Workflow Procedures" (with Mark Burstein et al.), presented at the Workshop on Semantics for the Rest of Us: Variants of Semantic Web Languages in the Real World, in conjunction with 8th International Semantic Web Conference (ISWC 2009), October 2009.
• "A new probabilistic plan recognition algorithm based on string rewriting" (with Christopher Geib and Robert P. Goldman), Proceedings of the Eighteenth International Conference on Automated Planning and Scheduling, September 2008.
• "Trapping Malicious Insiders in the SPDR Web" (with J. Thomas Haigh et al.), Proceedings of the 42nd Hawaii International Conference on System Sciences, January 2008.
• Older papers on programming languages and theory:
  • "Call-by-name, call-by-value, call-by-need and the linear lambda calculus" (with Martin Odersky, Philip Wadler and David N. Turner), Theoretical Computer Science, 228:1-2, 1999.
  • "The call-by-need lambda calculus" (with Martin Odersky and Philip Wadler), Journal of Functional Programming 8:3, May 1998.
  • "A pattern-based approach for a flexible, self-extending document system" (with Christian Salzmann), 24th International Conference and Exhibition on Technology of Object-Oriented Languages and Systems, September 1997.

Education

Dr.,Rer.,Nat., Computer Science, University of Karlsruhe, 1996

M.E., Computer Science, Tulane University, 1993

B.S., Computer Science and Mathematics, Tulane University, 1992