An Adaptive Unified Differential Evolution Algorithm for Global Optimization
In this paper, we propose a new adaptive unified differential evolution algorithm for single-objective global optimization. Instead of the multiple mutation strate- gies proposed in conventional differential evolution algorithms, this algorithm employs a single equation unifying multiple strategies into one expression. It has the virtue of mathematical simplicity and also provides users the flexibility for broader exploration of the space of mutation operators. By making all control parameters in the proposed algorithm self-adaptively evolve during the process of optimization, it frees the application users from the burden of choosing appro- priate control parameters and also improves the performance of the algorithm. In numerical tests using thirteen basic unimodal and multimodal functions, the proposed adaptive unified algorithm shows promising performance in compari- son to several conventional differential evolution algorithms.
- Research Organization:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- Accelerator & Fusion Research Division
- DOE Contract Number:
- DE-AC02-05CH11231
- OSTI ID:
- 1163660
- Report Number(s):
- LBNL-6853E
- Journal Information:
- Applied Soft Computing, Journal Name: Applied Soft Computing
- Country of Publication:
- United States
- Language:
- English
Similar Records
Accelerating Markov chain Monte Carlo simulation by differential evolution with self-adaptive randomized subspace sampling
Limited‐memory adaptive snapshot selection for proper orthogonal decomposition