A parallel implementation of the covariance matrix adaptation evolution\n strategy

2018/05/28 by Najeeb Alam Khan, Khan, Najeeb
Computer Science · Engineering · #Advanced Algorithms and Applications #Evolutionary Algorithms and Applications #FOS: Computer and information sciences #FOS: Mathematics #Magnetic Bearings and Levitation Dynamics #Metaheuristic Optimization Algorithms Research #Neural and Evolutionary Computing (cs.NE) #Optimization and Control (math.OC)

paper · pdf · doi:10.48550/arxiv.1805.11201

openalex publication_date 2018/05/28 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28

Abstract

In many practical optimization problems, the derivatives of the functions to\nbe optimized are unavailable or unreliable. Such optimization problems are\nsolved using derivative-free optimization techniques. One of the\nstate-of-the-art techniques for derivative-free optimization is the covariance\nmatrix adaptation evolution strategy (CMA-ES) algorithm. However, the\ncomplexity of CMA-ES algorithm makes it undesirable for tasks where fast\noptimization is needed. To reduce the execution time of CMA-ES, a parallel\nimplementation is proposed, and its performance is analyzed using the benchmark\nproblems in PythOPT optimization environment.\n

Related

• A Computationally Efficient Limited Memory CMA-ES for Large Scale\n Optimization
• CMA-ES for Hyperparameter Optimization of Deep Neural Networks
• Massively parallel CMA-ES with increasing population
• Covariance Matrix Adaptation Evolution Strategy Assisted by Principal Component Analysis
• Experimental Comparisons of Derivative Free Optimization Algorithms