Skip to main content

Edward Katz

Associate Teaching Professor - Silicon Valley

Contact

Eed.katz@northeastern.edu

Office Location

Northeastern University - Silicon Valley
6024 Silver Creek Valley Rd
San Jose, CA 95138

Mailing Address

Northeastern University
ATTN: Ed Katz, 6024 Silver Creek Valley Rd
360 Huntington Avenue
Boston, MA 02115

Research Interests

  • Autonomous mobile robot software systems
  • Computational intelligence (primarily Fuzzy Logic)
  • Smart environments and software agents
  • Rule-based, knowledge-based, and model-based multi-agent systems
  • Computation
  • Traditional Artificial Intelligence (GOFAI).

Education

  • Visiting Scholar, Robotics Laboratory, Stanford University
  • PhD in Computer Science, University of Louisiana at Lafayette
  • MS in Computer Science, University of Missouri-Columbia
  • BS in Mathematics, Purdue University

Biography

Edward Katz is an associate teaching professor in the M.S. Computer Science and Align programs within the Khoury College of Computer Sciences at Northeastern University’s Silicon Valley campus. His career began in Palo Alto, California, where he was a Senior Software Researcher at Hewlett-Packard Laboratories. He then spent ten years at Carnegie Mellon University’s Silicon Valley campus, where he was both a Senior Systems Scientist and an associate professor in software engineering. At Carnegie Mellon, he mentored teams of master’s students for the Entrepreneurial and Innovation Course. Between 2016 to 2017, he was a visiting associate professor of computer science at Loyola Marymount University.

In 1995, when Katz was invited to be a visiting scholar at the Stanford University Computer Science Department’s Robotics Laboratory, he collaborated with the legendary AI Pioneer professor Nils J. Nilsson. That collaboration resulted in Katz developing the Fuzzy Teleo-Reactive extended paradigm for autonomous robot agent control, an extension of Nilsson’s Teleo-Reactive pioneering work.

He has published technical papers in international conferences and has been awarded two United States Patents. His professional memberships include: Senior Member, Association for the Advancement of Artificial Intelligence (AAAI); Senior Member, Association for Computing Machinery (ACM); and Senior Member, Institute of Electrical and Electronics Engineers (IEEE). In 2018, he became the IEEE Robotics and Automation Society Chapter Chair for the entire San Francisco Bay Area region.

 

 

Recent Publications

Generating, Executing, and Monitoring Plans with Goal-Based Utilities in Continuous Domains

Langley, P., Choi, D., Barley, M., Meadows, B., & Katz, E. P. (2017). Generating, executing, and monitoring plans with goal-based utilities in continuous domains. Proceedings of the Fifth Annual Conference on Cognitive Systems. Troy, NY.

Abstract

Like humans, autonomous agents must operate in continuous physical settings which involve competing objectives that change over time. The PUG architecture addresses these issues by combining relational and quantitative descriptions of states, associating numeric utilities to symbolic goals, and using mental simulation to evaluate alternative plans. In this paper, we report extensions to PUG that interleave plan generation with execution and monitoring. We describe each of these processes, how they interact, and how they lead to plan revision when anomalies arise. We demonstrate the extended architecture’s behavior in a continuous physical domain, including four classes of scenarios in which unexpected developments lead the agent to revise its plans before continuing execution. In closing, we review work on related topics and discuss avenues for future research.

Goals, utilities, and mental simulation in continuous planning

Langley, P., Barley, M., Meadows, B., Choi, D., & Katz, E. P. (2016). Goals, utilities, and mental simulation in continuous planning. Proceedings of the Fourth Annual Conference on Cognitive Systems. Evanston, IL.

Abstract

Like humans, autonomous agents will need to operate in physical settings that involve competing objectives which may be incompatible and vary in importance over time. They will also need to reason about both qualitative relations and quantitative attributes to produce behavior that is appropriate to the situation. In this paper, we report PUG, a problem-solving architecture that combines conceptual inference with plan generation, encodes both relational and quantitative content, and integrates symbolic goals with numeric utilities. Mental simulation plays a key role in evaluating operators, guiding search, and determining when a plan is acceptable. We describe the system’s representational assumptions and the mechanisms that operate over them, after which we demonstrate its behavior in settings that involve conflicting goals with utilities that vary over time. In closing, we discuss related work and outline our plans for future research.