A New Parallel Cuckoo Flower Search Algorithm for Training Multi-Layer Perceptron

Abstract

This paper introduces a parallel meta-heuristic algorithm called Cuckoo Flower Search (CFS). This algorithm combines the Flower Pollination Algorithm (FPA) and Cuckoo Search (CS) to train Multi-Layer Perceptron (MLP) models. The algorithm is evaluated on standard benchmark problems and its competitiveness is demonstrated against other state-of-the-art algorithms. Multiple datasets are utilized to assess the performance of CFS for MLP training. The experimental results are compared with various algorithms such as Genetic Algorithm (GA), Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), Evolutionary Search (ES), Ant Colony Optimization (ACO), and Population-based Incremental Learning (PBIL). Statistical tests are conducted to validate the superiority of the CFS algorithm in finding global optimum solutions. The results indicate that CFS achieves significantly better outcomes with a higher convergence rate when compared to the other algorithms tested. This highlights the effectiveness of CFS in solving MLP optimization problems and its potential as a competitive algorithm in the field.

Comparing Metaheuristic Search Techniques in Addressing the Effectiveness of Clustering-Based DDoS Attack Detection Methods

“…According to Maddaiah et al [36], CSA applies three rules that are based on the breeding behavior of some cuckoo birds. The first rule is that every cuckoo lays one egg and places it in a random nest; the second rule is that the nest with the highest fitness proceeds to the next generation [37]. The third rule is that the probability of finding an egg is in the range of 0 and 1 using the 'pa' parameter [36,37].…”

“…The first rule is that every cuckoo lays one egg and places it in a random nest; the second rule is that the nest with the highest fitness proceeds to the next generation [37]. The third rule is that the probability of finding an egg is in the range of 0 and 1 using the 'pa' parameter [36,37]. As brought by Salgotra [37], incorporating these rules is achieved by using the Levy flight to generate a new solution as presented below.…”

“…The third rule is that the probability of finding an egg is in the range of 0 and 1 using the 'pa' parameter [36,37]. As brought by Salgotra [37], incorporating these rules is achieved by using the Levy flight to generate a new solution as presented below. Based on Sicuaio et al [38], Lévy (λ) is the Levy distribution in performing random walks.…”

“…Based on Sicuaio et al [38], Lévy (λ) is the Levy distribution in performing random walks. X t i is considered the previous solution, Ꚛ is regarded as entry-wise multiplication, and α is the step size for which in most cases the value of 1 is used [37].…”

“…Global pollination is achieved through Levy distribution [39,40]. According to Salgotra et al [37], the third rule states that flower consistency is proportionate to the likeness of two flowers, which is represented as the reproduction probability. The fourth rule refers to the use of pollination approximation.…”