Denoising autoencoder genetic programming (DAE-GP) is a novel neural network-based estimation of distribution genetic programming approach that uses denoising autoencoder long short-term memory networks as a probabilistic model to replace the standard mutation and recombination operators of genetic programming. At each generation, the idea is to capture promising properties of the parent population in a probabilistic model and to use corruption to transfer variations of these properties to the offspring. This work studies the influence of corruption and sampling steps on search. Corruption partially mutates candidate solutions that are used as input to the model, whereas the number of sampling steps defines how often we re-use the output during model sampling as input to the model. We study the generalization of the royal tree problem, the Airfoil problem, and the Pagie-1 problem, and find that both corruption strength and the number of sampling steps influence exploration and exploitation in search and affect performance: exploration increases with stronger corruption and lower number of sampling steps. The results indicate that both corruption and sampling steps are key to the success of the DAE-GP: it permits us to balance the exploration and exploitation behavior in search, resulting in an improved search quality. However, also selection is important for exploration and exploitation and should be chosen wisely.

去噪自编码器遗传规划（DAE-GP）是一种新颖的基于神经网络的分布遗传规划估计方法，它使用去噪自编码器长短期记忆网络作为概率模型来取代遗传规划的标准变异和重组算子。在每一代，我们的想法都是在概率模型中捕获父代群体的有前途的特性，并利用腐败将这些特性的变化传递给后代。这项工作研究了腐败和抽样步骤对搜索的影响。损坏会部分改变用作模型输入的候选解决方案，而采样步骤的数量定义了我们在模型采样期间重新使用输出作为模型输入的频率。我们研究了 Royal Tree 问题、Airfoil 问题和 Pagie-1 问题的泛化，发现腐败强度和采样步骤数都会影响搜索中的探索和利用，并影响性能：腐败越强，探索就越多。采样步数。结果表明，腐败和采样步骤是 DAE-GP 成功的关键：它使我们能够平衡搜索中的探索和利用行为，从而提高搜索质量。然而，选择对于勘探和开发也很重要，应该明智地选择。