Incremental evolution of complex general behavior
Incremental evolution of complex general behavior
复制标题
DOI：
10.1177/105971239700500305
复制
发表时间：
1997-12-01
影响因子：
1.6
通讯作者：
Miikkulainen, R
中科院分区：
文献类型：
--
作者：
Gomez, F;Miikkulainen, R
几位研究人员已经演示了如何通过神经进化(即用遗传算法进化神经网络)来学习复杂的动作序列。然而，复杂的一般行为，如躲避捕食者或躲避障碍物，与特定环境无关，事实证明很难进化。通常，系统会发现机械的策略，如来回移动，这些策略有助于代理应对，但不是很有效，看起来不可信，也不适用于新环境。问题是，对于进化系统来说，一般策略太难直接发现了。本文提出了一种方法，其中这种复杂的一般行为是通过增量学习的，从更简单的行为开始，逐渐使任务更具挑战性和一般性。任务转移通过Delta编码的连续阶段(即进化修改)来实现，这使得即使是收敛的种群也能适应新的任务。在随机、动态的捕食任务中对该方法进行了测试，并与直接进化算法进行了比较。渐进式方法发展出更有效和更通用的行为，还应该扩展到更困难的任务。
Several researchers have demonstrated how complex action sequences can be learned through neuroevolution (i.e., evolving neural networks with genetic algorithms). However, complex general behavior such as evading predators or avoiding obstacles, which is not tied to specific environments, turns out to be very difficult to evolve. Often the system discovers mechanical strategies, such as moving back and forth, that help the agent cope but are not very effective, do not appear believable, and do not generalize to new environments. The problem is that a general strategy is too difficult for the evolution system to discover directly. This article proposes an approach wherein such complex general behavior is learned incrementally, by starting with simpler behavior and gradually making the task more challenging and general The task transitions are implemented through successive stages of Delta coding (i.e., evolving modifications), which allows even converged populations to adapt to the new task. The method is tested in the stochastic, dynamic task of prey capture and is compared with direct evolution. The incremental approach evolves more effective and more general behavior and should also scale up to harder tasks.