Evolving Deep Neural Network by Customized Moth Flame Optimization Algorithm for Underwater Targets Recognition

Mohammad Khishe; Mokhtar Mohammadi; Tarik A. Rashid; Hoger Mahmud; Seyedali Mirjalili

Home/Archive/arXiv

arXiv · 2303.00922

Evolving Deep Neural Network by Customized Moth Flame Optimization Algorithm for Underwater Targets Recognition

Mohammad Khishe·Mokhtar Mohammadi·Tarik A. Rashid·Hoger Mahmud·Seyedali Mirjalili

View original source ↗Read source document ↗Export metadata ↓

Abstract

This chapter proposes using the Moth Flame Optimization (MFO) algorithm for finetuning a Deep Neural Network to recognize different underwater sonar datasets. Same as other models evolved by metaheuristic algorithms, premature convergence, trapping in local minima, and failure to converge in a reasonable time are three defects MFO confronts in solving problems with high-dimension search space. Spiral flying is the key component of the MFO as it determines how the moths adjust their positions in relation to flames; thereby, the shape of spiral motions can regulate the transition behavior between the exploration and exploitation phases. Therefore, this chapter investigates the efficiency of seven spiral motions with different curvatures and slopes in the performance of the MFO, especially for underwater target classification tasks. To assess the performance of the customized model, in addition to benchmark Sejnowski & Gorman's dataset, two experimental sonar datasets, i.e., the passive sonar and active datasets, are exploited. The results of MFO and its modifications are compared to four novel nature-inspired algorithms, including Heap-Based Optimizer (HBO), Chimp Optimization Algorithm (ChOA), Ant Lion Optimization (ALO), Stochastic Fractals Search (SFS), as well as the classic Particle Swarm Optimization (PSO). The results confirm that the customized MFO shows better performance than the other state-of-the-art models so that the classification rates are increased 1.5979, 0.9985, and 2.0879 for Sejnowski & Gorman, passive, and active datasets, respectively. The results also approve that time complexity is not significantly increased by using different spiral motions.

Explore related subjects

cs.NE

Keep this discovery

BibTeX RIS

Mohammad Khishe, Mokhtar Mohammadi, Tarik A. Rashid, Hoger Mahmud, Seyedali Mirjalili. 2023-02-16. Evolving Deep Neural Network by Customized Moth Flame Optimization Algorithm for Underwater Targets Recognition. https://arxiv.org/abs/2303.00922

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Breaking the Central Bias: Spatially Partitioned Experts for Coordinate-Based Neuroevolution

Evolvable-Substrate HyperNEAT (ES-HyperNEAT), a bio-inspired indirect encoding that determines neuron placement and connection weights from spatial coordinates, exhibits a failure mode on MNIST as a diagnostic benchmark. Because input pixels map to a coordinate space centered at the origin, evolved networks converge on a small central cluster of input pixels, a spatial-concentration bias; prior work observed only 21% mean accuracy in this regime. Is this bias an optimization artifact or an architectural ceiling? Inspired by Mixture-of-Experts (MoE) principles, we partition the input into non-overlapping spatial segments, each assigned to a separately evolved specialist network. With 13 such experts, this design reaches 43% mean accuracy, a 106% relative improvement over the baseline. The architectural gain does not depend on data-driven aggregation: equal-weighted averaging, which uses no validation data, already yields a 70% improvement; the gain comes from partitioning, not the weighting. Receptive-field analysis shows the mechanism: partitioning forces evolution to discover features across the entire image, expanding active pixel coverage from 4% to 79%. Absolute accuracy stays below gradient-trained baselines, but the relative gain points to central bias, not the evolutionary search. Two tools are designed to generalize beyond MNIST: a receptive-field diagnostic for silent input-coverage collapse, and a spatial-partitioning remedy that restores coverage.

cs.NE↗

arXiv2026

A Bio-Plausible Visual Neural Network for Locust-Inspired Collision Perception

Locust visual systems have long served as an important biological paradigm for studying looming perception and collision avoidance. Numerous computational models have successfully reproduced the selective responses of Lobula Giant Movement Detector (LGMD) neurons to approaching objects, thereby emulating the fundamental functionality of the biological system. However, existing models remain limited in biological plausibility and robustness when operating in complex and dynamic visual environments. To address these limitations, we propose a biologically plausible neural network for locust-inspired looming detection. The proposed framework incorporates a spatially isotropic sampling strategy that mimics the ommatidial organization of the locust compound eye, a population-voting mechanism inspired by population coding in biological neural systems, and leaky integrate-and-fire neuronal dynamics to replace conventional sigmoid-based membrane activation. Systematic experiments on synthetic stimuli, laboratory sequences, and real-world driving scenarios demonstrate that the proposed model improves robustness under challenging visual conditions while preserving computational efficiency and enhancing biological fidelity. These results highlight the potential of biologically grounded neural computation for robust and efficient collision perception.

cs.NE↗

arXiv2026

Structural Fusion of Bayesian Networks with Limited Treewidth Using Genetic Algorithms

This paper introduces an evolutionary computation approach for consensus in structural Bayesian Network (BN) fusion under the constraint of limited treewidth. The consensus BN aims to reconcile multiple input BNs into a single one that retains key structural features present in the original networks. Treewidth, a graph-based parameter associated with computationally tractable inference, is utilized to restrict the complexity of the resulting network. A genetic algorithm is proposed to look for a BN that codifies as much information about the unrestricted fusion as possible while ensuring the treewidth restriction. Experimental evaluation demonstrates the genetic algorithm's ability to obtain consensus BNs with limited treewidth, providing a valuable tool for aggregating information from diverse sources while returning a computationally actionable model.

cs.NE↗