Metaheuristics Meets Statistics: Cuckoo Search for Designs, Inference, and Data Geometry

Abstract Nature-inspired metaheuristic algorithms have grown by leaps and bounds in the last three decades, and there are now many new and powerful algorithms with increasing ability to tackle all kinds of optimization problems across disciplines. This paper shows the diverse and new applications of a typical nature-inspired metaheuristic algorithm called Cuckoo or Cuckoo Search (CS). As examples, we demonstrate applications of the CS algorithm to find optimal exact designs for estimating model parameters in a nonlinear regression model, tackle complicated estimation problems in bioinformatics, and how CS efficiently constructs principal curves to visualize and analyze a large cloud of data.