协同共进化 (CC)

class pypop7.optimizers.cc.cc.CC(problem, options)[源代码]

协同共进化 (Cooperative Coevolution, CC)。

这是所有 CC 类的抽象类。请使用其任何实例化的子类来优化您手头的黑盒问题。

注意

CC 使用分解策略来缓解大规模黑盒优化中的维度灾难问题。关于收敛性分析，请参考 [Panait et al., 2008, JMLR]；关于 Schmidhuber 和/或 Miikkulainen 实验室最前沿的神经进化应用，可参考 [Gomez et al.,, 2008, JMLR] 等文献。

参数:

  • problem (dict) –

    问题参数，包含以下通用设置 (键)

    • 'fitness_function' - 需要被最小化的目标函数 (func),

    • 'ndim_problem' - 维度数量 (int),

    • 'upper_boundary' - 搜索范围的上边界 (array_like),

    • 'lower_boundary' - 搜索范围的下边界 (array_like).

  • options (dict) –

    优化器选项，包含以下通用设置 (键)

    • 'max_function_evaluations' - 函数评估的最大次数 (int, 默认: np.inf),

    • 'max_runtime' - 允许的最大运行时间 (float, 默认: np.inf),

    • 'seed_rng' - 随机数生成器的种子，需要明确设置 (int);

    以及以下特定设置 (key)

    • “n_individuals” - 个体/样本的数量，也称为种群大小（int，默认值：100）。

参考文献

Gomez, F., Schmidhuber, J. and Miikkulainen, R., 2008. Accelerated neural evolution through cooperatively coevolved synapses. Journal of Machine Learning Research, 9(31), pp.937-965. https://www.jmlr.org/papers/v9/gomez08a.html

Panait, L., Tuyls, K. and Luke, S., 2008. Theoretical advantages of lenient learners: An evolutionary game theoretic perspective. Journal of Machine Learning Research, 9, pp.423-457. https://jmlr.org/papers/volume9/panait08a/panait08a.pdf

Schmidhuber, J., Wierstra, D., Gagliolo, M. and Gomez, F., 2007. Training recurrent networks by evolino. Neural Computation, 19(3), pp.757-779. https://direct.mit.edu/neco/article-abstract/19/3/757/7156/Training-Recurrent-Networks-by-Evolino

Gomez, F.J. and Schmidhuber, J., 2005, June. Co-evolving recurrent neurons learn deep memory POMDPs. In Proceedings of Annual Conference on Genetic and Evolutionary Computation (pp. 491-498). ACM. https://dl.acm.org/doi/10.1145/1068009.1068092

Fan, J., Lau, R. and Miikkulainen, R., 2003. Utilizing domain knowledge in neuroevolution. In International Conference on Machine Learning (pp. 170-177). https://www.aaai.org/Library/ICML/2003/icml03-025.php

Potter, M.A. and De Jong, K.A., 2000. Cooperative coevolution: An architecture for evolving coadapted subcomponents. Evolutionary Computation, 8(1), pp.1-29. https://direct.mit.edu/evco/article/8/1/1/859/Cooperative-Coevolution-An-Architecture-for

Gomez, F.J. and Miikkulainen, R., 1999, July. Solving non-Markovian control tasks with neuroevolution. In Proceedings of International Joint Conference on Artificial Intelligence (pp. 1356-1361). https://www.ijcai.org/Proceedings/99-2/Papers/097.pdf

Moriarty, D.E. and Mikkulainen, R., 1996. Efficient reinforcement learning through symbiotic evolution. Machine Learning, 22(1), pp.11-32. https://link.springer.com/article/10.1023/A:1018004120707

Moriarty, D.E. and Miikkulainen, R., 1995. Efficient learning from delayed rewards through symbiotic evolution. In International Conference on Machine Learning (pp. 396-404). Morgan Kaufmann. https://www.sciencedirect.com/science/article/pii/B9781558603776500566

Potter, M.A. and De Jong, K.A., 1994, October. A cooperative coevolutionary approach to function optimization. In International Conference on Parallel Problem Solving from Nature (pp. 249-257). Springer, Berlin, Heidelberg. https://link.springer.com/chapter/10.1007/3-540-58484-6_269

CC