Blackbox Optimization: Implications Of SEARCH

The SEARCH (Search Envisioned As Relation & Class Hierarchizing) framework developed elsewhere (Kargupta, 1995; Kargupta & Goldberg, 1996a; Kargupta & Goldberg, 1996b) offered an alternate perspective toward blackbox optimization---optimization in presence of little domain knowledge. The SEARCH framework investigates the conditions essential for transcending the limits of random enumerative search using a framework developed in terms of relations, classes and partial ordering. The primary motivation comes from the observation that sampling in blackbox optimization is essentially an inductive process (Michalski, 1983) and in absence of any relation among the members of the search space, induction is no better than enumeration (Watanabe, 1969). This paper presents the main implications of SEARCH. The lessons from SEARCH are used to develop a comprehensive approach to identify the role of algorithm, problem, and the user in blackbox optimization. The computational benefit of implicit para..