A Learning—Based Particle Swarm Optimizer for Solving Mathematical Combinatorial Problems

Olivares, Rodrigo; Soto, Ricardo; Crawford, Broderick; Ríos, Víctor; Olivares, Pablo; Ravelo, Camilo; Medina, Sebastian; Nauduan, Diego

doi:10.3390/axioms12070643

Cited by 4 publications

(2 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The synergy between advanced machine learning techniques and metaheuristics is pivotal in crafting solutions that effectively address the sophisticated and ever-evolving landscape of cyber threats. Notably, research such as [ 19 ] emphasizes the utility of integrating Q-Learning with Particle Swarm Optimization for the resolution of combinatorial problems, marking a significant advancement over traditional PSO methodologies. The approach not only enhances solution quality but also exemplifies the effectiveness of learning-based hybridizations in the broader context of swarm intelligence algorithms, providing a novel and adaptable methodology for tackling optimization challenges.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Enhancing the Efficiency of a Cybersecurity Operations Center Using Biomimetic Algorithms Empowered by Deep Q-Learning

Olivares,

Salinas,

Ravelo

et al. 2024

Biomimetics

Self Cite

View full text Add to dashboard Cite

In the complex and dynamic landscape of cyber threats, organizations require sophisticated strategies for managing Cybersecurity Operations Centers and deploying Security Information and Event Management systems. Our study enhances these strategies by integrating the precision of well-known biomimetic optimization algorithms—namely Particle Swarm Optimization, the Bat Algorithm, the Gray Wolf Optimizer, and the Orca Predator Algorithm—with the adaptability of Deep Q-Learning, a reinforcement learning technique that leverages deep neural networks to teach algorithms optimal actions through trial and error in complex environments. This hybrid methodology targets the efficient allocation and deployment of network intrusion detection sensors while balancing cost-effectiveness with essential network security imperatives. Comprehensive computational tests show that versions enhanced with Deep Q-Learning significantly outperform their native counterparts, especially in complex infrastructures. These results highlight the efficacy of integrating metaheuristics with reinforcement learning to tackle complex optimization challenges, underscoring Deep Q-Learning’s potential to boost cybersecurity measures in rapidly evolving threat environments.

show abstract

Section: Related Workmentioning

confidence: 99%

“…DQL is governed by a loss function according to Equation (19), which measures the discrepancy between the estimated Q and target values:…”

Section: Reinforcement Learningmentioning

confidence: 99%