Many CISAC projects involve decision makers who have to respond to a threat quickly in spite of uncertain evidence that it is really there. Surprise attack, the stability of nuclear arsenals, monitoring an arms agreement, intervening against genocide, distinguishing epidemics from bioterrorism, and intercepting a dangerous asteroid -- these all require detecting a threat quickly but avoiding a false alarm. Most of our work has involved assembling and judging the evidence of the threat, but further issues arise around the next step: what to do given the evidence. These problems become more subtle when the decision is a "strategic" one, i.e., made against an adversary who knows you are watching. The talk will illustrate the use of game theory to come to some counterintuitive conclusions.

Reading available in IIS library, 5th floor, Encina Hall East.