=Paper=
{{Paper
|id=None
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-841/talkQuirk.pdf
|volume=Vol-841
}}
==None==
<pdf width="1500px">https://ceur-ws.org/Vol-841/talkQuirk.pdf</pdf>
<pre>
        Deterrence Analysis: AI and Cognitive Science As Necessary
                                Ingredients

                                          Dr. Michelle Quirk

                                             Project Scientist
                         Basic and Applied Research Office InnoVision Directorate
                                 National Geospatial-Intelligence Agency


Abstract                                                 Biographical Sketch

The concept of deterrence can be defined as              Michelle Quirk is a project scientist in the Basic
the display of possible threats by one party to          and Applied Research Office, InnoVision
convince another party to refrain from initiating        Directorate, National Geospatial-Intelligence
certain courses of action. The most common               Agency (NGA). She has a career that spans over
deterrent is that of a threat that convinces the         25 years in the areas of applied computer
adversary not to carry out intended actions              science and computational mathematics.
because of costly consequences.
 In a nutshell, deterrence analysis comprises            Michelle began her research activities in
these three factors:                                     modeling rock mechanics and continued with
     • The benefits of a course of action                numerical methods for solving partial
     • The costs of a course of action                   differential equations, where she pioneered the
     • The consequences of restraint (i.e.,              work on infinite elements method for the
         costs and benefits of not taking the            Maxwell's Equations. As a scientist at Los
         course of action we seek to deter).             Alamos National laboratory, Michelle developed
Deterrence analysis aims to create scores based          extensions to JPEG-2000 Standard with
on these three elements. Often these exercises           applications to hyper-spectral imagery analysis
are static in nature and are conducted for a             and processing.
special problem, without automation or
sensible reuse of previous results.           One        Prior to joining NGA, Michelle was a
preferred method was to create typical colored           computational scientist at the Defense Threat
tables (HIGH, MEDIUM, LOW) of scores and                 Reduction Agency, where she worked on
risks. These tables were compiled mostly                 strategic deterrence assessments and decision
manually. A more elegant approach was game               analysis under uncertainty with non-
theory. In this talk we discuss briefly behavioral       probabilistic, soft metrics.
game theory and its suitability to deterrence
analysis. Further, we propose a computational            Michelle earned a M.S. in computational and
framework that that has an Artificial                    applied mathematics (1994) and a Ph.D. in
Intelligence foundation and employs cognitive            mathematics (2003), from the University of
sciences in the design and as means to achieve           Texas at Austin.
a permeating validation.
We will close the talk with a list of deterrence
open questions and an example of their
refinement, as a first step to create a true
computational engine.

</pre>