Bolouki, S; Manshaei, MH; Ravanmehr, V; Nedić, A; Başar, T
abstract
Group testing offers a cost/time-beneficial method to identify all, but few, infected individuals (defective items in general) among a large set of individuals (items). In a group testing scheme, a series of tests are performed on groups of individuals rather than single individuals. A test on a group determines whether the group contains at least one infected individual. This paper investigates the classical group testing problem from a game-theoretic perspective, where every individual, once called for a test, decides to comply with or defy the call. In this framework, an individual's decision is driven by his knowledge of his well-being, that is healthy or infected. This leads to the so-called group testing game which is formulated in this work. Some simplified versions of the general game as a team game is then presented and analyzed, that result in some novel, generalized group testing problems to be addressed in future work.
