Abstract

Neural architecture search (NAS) is a rapidly growing field which focuses on the automated design of neural network architectures. Genetic algorithms (GAs) have been predominantly used for evolving neural network architectures. Genetic programming (GP), a variation of GAs that work in the program space rather than a solution space, has not been as well researched for NAS. This paper aims to contribute to the research into GP for NAS. Previous research in this field can be divided into two categories. In the first each program represents neural networks directly or components and parameters of neural networks. In the second category each program is a set of instructions, which when executed, produces a neural network. This study focuses on this second category which has not been well researched. Previous work has used grammatical evolution for generating these programs. This study examines canonical GP for neural network design (GPNND) for this purpose. It also evaluates a variation of GP, iterative structure-based GP (ISBGP) for evolving these programs. The study compares the performance of GAs, GPNND and ISBGP for image classification and video shorts creation. Both GPNND and ISBGP were found to outperform GAs, with ISBGP producing better results than GPNND for both applications. Both GPNND and ISBGP produced better results than previous studies employing grammatical evolution on the CIFAR-10 dataset.

中文翻译：

---

用于图像分类和视频短片创建的 CNN 模型自动设计的遗传编程方法

摘要

神经架构搜索（NAS）是一个快速发展的领域，专注于神经网络架构的自动化设计。遗传算法（GA）主要用于进化神经网络架构。遗传编程 (GP) 是 GA 的一种变体，在程序空间而不是解决方案空间中工作，对于 NAS 的研究还不够深入。本文旨在为 NAS 的 GP 研究做出贡献。先前该领域的研究可分为两类。在第一个程序中，每个程序直接代表神经网络或神经网络的组件和参数。在第二类中，每个程序都是一组指令，执行时会产生一个神经网络。本研究主要针对尚未得到充分研究的第二类。之前的工作已经使用语法演化来生成这些程序。为此，本研究检查了神经网络设计的规范 GP (GPNND)。它还评估 GP 的变体，即基于迭代结构的 GP (ISBGP)，以改进这些程序。该研究比较了 GA、GPNND 和 ISBGP 在图像分类和视频短片创作方面的性能。研究发现 GPNND 和 ISBGP 的性能都优于 GA，对于这两种应用，ISBGP 都比 GPNND 产生更好的结果。GPNND 和 ISBGP 都比之前在 CIFAR-10 数据集上使用语法进化的研究产生了更好的结果。