Ugur Kuter
Ph.D., Computer Science, University of Maryland, College Park, MD, 2006.
B.S. / M.S., Computer Engineering, Middle East Technical University, Ankara, 1998 / 2000.
Dr. Kuter is a SIFT principal scientist specializing in Artificial Intelligence, studying intelligence and intelligent behavior, human and computational, in both individuals and groups. He is particularly interested in understanding and developing intelligent systems that can observe, think, and (inter-)act with other such systems, including humans. A large part of Dr. Kuter's research is founded in Artificial Intelligence, particularly around: Automated planning, Machine learning, Reasoning with / about uncertainty, Knowledge representations, Graph analysis, and Game Theory.
Currently, he is working on the applications of the above research topics in decentralized automated planning and learning, adversarial reasoning and planning, explainable and transparent planning, team social interactions, social networks, synthetic biology, cyber resilience, and software program analysis. He is currently leading SIFT efforts in the following projects:
- Hierarchical Operations Management and Execution Reasoner (HOMER). U.S. Army Combat Capabilities Development Command Advanced Teaming (ATeam)
- Assessing SBS Research through Automated Paper Comprehension and Analysis (A+). DARPA Systematizing Confidence in Open Research and Evidence (SCORE)
- Collaboration Facilitator Agent Study (CFAS). DARPA I2O Seedling.
Currently, Dr. Kuter is also leading technical areas in the following projects:
- Artificial Social Intelligence for Successful Teams: Architecture for Natural Teamwork (ASIST:ANT). DARPA Artificial Social Intelligence for Successful Teams (ASIST)
- System for Counterfactual Human Network Evaluation of Individual Differences, Errors, and Residuals (SCHNEIDER). DARPA Teaching AI to Leverage Overlooked Residuals (TAILOR).
Previously, Dr. Kuter has led and successfully completed SIFT's effort in AFRL Distributed Operations (DistrO) program, sub-contracting to Lockheed Martin CO on our LaPlata project and was a key personnel on DARPA Collection and Monitoring via Planning for Active Situational Scenarios (COMPASS) program, subcontracting to BBN Technologies on our Gray-zone Recognition of Adversary Intent via Learning (GRAIL) project. He has finished U.S.Navy SBIR MPRIMUS: Mission Planning for Resources in Integrated Mixed Undersea System and Army SBIR entitled VANESSA (Virtual Analysis Networks and Explanations for Social Sensing Analysis).He has contributed to SIFT efforts to DARPA SD2 (Synergy Discovery and Design) in the programs early stages.
Previously, Dr. Kuter has led and completed our Machine Self-Confidence elicitation efforts in SIFT's AFRL contract, entitled "Self-Confidence in Automation and our ONR project, entitled "HACKAR: Helpful Advice for Code Knowledge and Attack Resilience," for the "Computational Methods for Decision Making" program. He was a key personnel on the SIFT efforts on the DARPA's "Mission-oriented Resilient Clouds (MRC)" and "Collaborative Operations in Denied Environment (CODE)" programs, developing formalisms and algorithms for probabilistic plan recognition, diverse planning, hierarchical plan repair, and plan-critiquing systems in highly dynamic mixed-initiative environments. He has also completed SIFT's mission planning, counter planning, and plan repair efforts in the DARPA CODE (Collaborative Operations in Denied Environments) and DARPA OBTW (Oh By The Way) programs.
Since his PhD disertation and aftgerwards, Dr. Kuter has been one of the primary contributors to the SHOP3 (Simple Hierarchical Ordered Planner v3) Hierarchical Task Network planning system, invented originally at the University of Maryland and have been maintained and extended at SIFT over 15 years now. SHOP3 has been downloaded thousans of times by researchers both from academia and industry over the years.
Dr. Kuter has been a regular reviewer/program committee member for several international conferences and journals, including IJCAI, AAAI, ICAPS, AAMAS, IEEE Intelligent Systems, Artificial Intelligence Journal, and Journal of Artificial Intelligence Research, IEEE Transactions on Knowledge and Data Engineering. He has more than 100 peer-viewed research publications, including several best papers. Dr. Kuter's work on "Social Trust based Web Service Composition" won the Best Paper Award at ISWC-09, one of the predominant conferences on Semantic Web. His paper "Using Probabilistic Confidence Models for Trust Inference in Web-Based Social Networks" was ranked in the top %5 of all accepted papers at AAAI-07, one of the top AI conferences. His work on the RFF automated-planning system has received the "Winner of the 2008 International Planning Competition -- Planning under Uncertainty, Full-Observable Planning Track" award. At the University of Maryland, he participated the University of Maryland's successful contributions to the DARPA's Integrated Learning and the Transfer Learning programs. For a complete list of Dr. Kuter's publications, please visit: http://sites.google.com/site/ukuter/