利用强化学习做函数全局优化问题是否可以呢？

有一种方法叫做Learning Automata，属于一种强化学习方法，可以与进化计算（如PSO、差分进化）、元胞自动机（CA）等算法结合起来，对某些问题进行优化，常见的是function optimization。可以参考下列文献：

1 Rezvanian A, Meybodi MR (2010) LACAIS: learning automata based cooperative artificial

immune system for function optimization. Contemporary computing. Springer, Berlin Heidelberg, pp 64–75

2 Hasanzadeh M, Meybodi MR, Ebadzadeh MM (2013) Adaptive cooperative particle swarm optimizer. Appl Intel 39:397–420

3 Moradabadi B, Beigy H (2014) A new real-coded Bayesian optimization algorithm based on a team of learning automata for continuous optimization. Genet Program Evolvable Mach 15:169–193

4 Mahdaviani M, Kordestani JK, Rezvanian A, Meybodi MR (2015) LADE: learning automata

based differential evolution. Int J Artif Intel Tools 24:1550023

5 Mirsaleh MR, Meybodi MR (2015) A learning automata-based memetic algorithm. Genet Program Evolvable Mach 16:399–4536

6 Moradabadi B, Meybodi MR (2016) Link prediction based on temporal similarity metrics using continuous action set learning automata. Phys A Stat Mech Appl 460:361–373

7 Vafashoar R, Meybodi MR (2016) Multi swarm bare bones particle swarm optimization with distribution adaption. Appl Soft Comput 47:534–552