October 4, 2002
Colloquium Speaker: Robert Fry
Mr. Robert Fry is on the Principal Professional Staff at the Applied Physics Laboratory where he has worked since 1979. He attended Carnegie Mellon University, Youngstown State University (BSEE), Johns Hopkins Medical School in biomedical engineering, Johns Hopkins University (MSEE), and the University of Maryland at College Park. He is currently in the Space Department working principally in the area of advanced concept development. He has received several awards from the Navy including the AEGIS Team Excellence Award and holds 4 patents in various areas. His professional interests include logic, probability and information theory, cybernetics, and system engineering.
That one can quantify questions and inquiry may seem remarkable. However, the late Prof. Richard T. Cox of the Physics department at Johns Hopkins University, developed the basis for a theory of questions. In this talk we develop the idea that questions can be regarded as logical quantities that can be mathematically manipulated and formalized through a theory of inquiry. The essential idea is that information is intrinsically relative to the observer desiring it. The theoretical and practical implications of a theory of inquiry are significant in that such a theory provides for a generalized theory information — one that includes both traditional information theory as developed by Shannon, and a new theory of information that is relevant to the design of optimal systems and architectures that operate cybernetically. In addition to providing simple geometrical constructs to help understand and visualize the notion of questions and inquiry, this talk discusses some uses of this theory. In particular, it is shown how neurons and the structure of the brain embrace many aspects of a theory of inquiry and that even a single neuron can be viewed as an optimal controller. An important practical engineering application of the theory is to support the design of new kinds of defense architectures that have intrinsic robustness to system uncertainties. In ballistic missile defense, it is shown that the discrimination problem and enemy countermeasures ostensibly deny the observing system from collapsing a probability function describing possible locations of the threat by way of analogy to quantum mechanics. The described theory's ties to physics are highlighted using various historical connections including the anticipatory work of Henry Poincaré. The talk ends with a discussion of the current state of the theory and the future steps necessary to develop a mature engineering formalism of inference and inquiry.