Smart Information Flow Technologies (SIFT) is a research and development consulting company specializing in Human Factors and Artificial Intelligence. SIFT's goal is to make the information flow between humans and technology better for both sides -- more efficient, productive, pleasant, and safer. In order to achieve this goal SIFT employs top engineers in the fields of Computer Science and Psychology who specialize in Human Computer Interaction, Interface Design, Human Performance, Artificial Intelligence, Network and Cybersecurity, and Politeness and Etiquette models.
Since our inception in 1999, SIFT personnel have extended the state of the art in a wide range of domains from commercial and military flight decks to DoD small unit operations and have authored well over a hundred papers documenting our many contributions to the state of the art in multiple fields.
Ms. Peggy Wu will present SIFT's NASA funded work at the AAAI spring symposium series (http://www.aaai.org/Symposia/Spring/sss15.php) in a paper entitled Mining for Psycho-social Dimensions through Socio-linguistics. She will discuss results from an ongoing three year effort developing non-intrusive linguistic techniques for detecting psycho-social states in a Mars exploration simulation. March 22-23, Palo Alto, CA.
Drs. Ugur Kuter, Mark Burstein, J. Benton, Dan Bryce, Jordan Thayer and Steve McCoy will present their article HACKAR: Helpful Advice for Code Knowledge and Attack Resiliencein the Emerging Applications track of the AAAI Innovative Applications of AI (IAAI-15) conference at Austin TX. This paper describes our HACKAR system, a novel combination of Java program analysis and automated learning and planning architecture to the domain of Java vulnerability analysis. The key feature of HACKAR is its ability to analyze Java programs at development-time, identifying vulnerabilities and ways to avoid them. A prepublication draft is available on request from Dr. Kuter. The published paper will be available in the AAAI archives as well as Dr. Kuter's web site.
Drs. Miller and Kuter recently traveled to the University of Colorado at Boulder and delivered a seminar on October 7th titled "Supervising Automation: Hierarchical Delegation for “Unmanned” Systems We Can Live With" to the Research and Engineering Center for Unmanned Vehicles (RECUV). We focused on presenting Playbook (R) research and integrative research on hierarchical network planning, learning and representations-- including some recent and innovative work on automation "self confidence" and its expression and use by human supervisors. The audience was interdisciplinary including representatives of aeronautics, atmospheric sciences, automation, controls, communications and signal processing disciplines.
Drs. Ugur Kuter and Robert Goldman are co-authors of the forthcoming article, "Plan Aggregation for Strong Cyclic Planning in Nondeterministic Domains," to appear in the prestigious journal, Artificial Intelligence (vol. 216, 2014). A prepublication draft is available on Dr. Goldman's homepage. The paper was written with researchers at the University of Maryland, Dr. Ron Alford, and Prof. Dana Nau, who have collaborated with SIFT on mulitple research projects (including, most recently, DARPA's OBTW program).