Jarkko KARI


Jonathan MILLS

James M. NYCE


Gabriela QUEIROZ


Philip WELCH

Evolvable hardware:
From successful applications to implications for the theory of computation

Abstract. Current work in the field of evolvable hardware can be split into the two related areas of evolutionary hardware design and adaptive hardware. While evolutionary hardware design is the use of evolutionary algorithms (EAs) for creating innovative (and sometimes patentable) physical designs, the goal of adaptive hardware is to endow physical systems with some adaptive characteristics in order to allow them to operate successfully in a changing environment or under presence of faults. The recent years of development of this field can be characterized as a continuous search for promising problems/applications from the point of view of evolutionary design. In this talk, various innovative circuits obtained using evolutionary algorithms will be presented and compared with the best-known conventional designs. Typical problems associated with the use of evolvable hardware will be identified and possible solutions will be sketched. In the second part of the talk, it will be shown that evolvable hardware can have an impact on the theory of computations. Evolved computational devices represent a distinctive class of devices which exhibits a specific combination of properties, not visible and studied in the scope of all computational devices up till now. The devices that belong to this class show the required behavior; however, in general, we do not understand how and why they perform the required computation. Therefore, nothing is known about the mapping between an abstract computational model and its physical implementation. The standard notion of computation and implementation developed in computer science as well as in cognitive science has become very problematic with the existence of evolved computational devices.