IEEE Trans Neural Netw Learn Syst. 2020 Apr;31(4):1242-1254. doi: 10.1109/TNNLS.2019.2919608. Epub 2019 Jun 20.
The performance of convolutional neural networks (CNNs) highly relies on their architectures. In order to design a CNN with promising performance, extensive expertise in both CNNs and the investigated problem domain is required, which is not necessarily available to every interested user. To address this problem, we propose to automatically evolve CNN architectures by using a genetic algorithm (GA) based on ResNet and DenseNet blocks. The proposed algorithm is completely automatic in designing CNN architectures. In particular, neither preprocessing before it starts nor postprocessing in terms of CNNs is needed. Furthermore, the proposed algorithm does not require users with domain knowledge on CNNs, the investigated problem, or even GAs. The proposed algorithm is evaluated on the CIFAR10 and CIFAR100 benchmark data sets against 18 state-of-the-art peer competitors. Experimental results show that the proposed algorithm outperforms the state-of-the-art CNNs hand-crafted and the CNNs designed by automatic peer competitors in terms of the classification performance and achieves a competitive classification accuracy against semiautomatic peer competitors. In addition, the proposed algorithm consumes much less computational resource than most peer competitors in finding the best CNN architectures.
卷积神经网络（CNN）的性能高度依赖于其架构。为了设计具有良好性能的 CNN，需要同时具备对 CNN 和研究问题领域的广泛专业知识，但并非每个感兴趣的用户都具备这些知识。为了解决这个问题，我们提出了一种使用基于 ResNet 和 DenseNet 块的遗传算法（GA）自动进化 CNN 架构的方法。所提出的算法在设计 CNN 架构时完全是自动的。具体来说，它不需要在开始之前进行预处理，也不需要在 CNN 方面进行后处理。此外，所提出的算法不需要用户具备关于 CNN、研究问题甚至 GA 的领域知识。所提出的算法在 CIFAR10 和 CIFAR100 基准数据集上与 18 种最先进的同行竞争者进行了评估。实验结果表明，在所提出的算法在分类性能方面优于手工制作的最先进的 CNN 和自动设计的 CNN，并且在与半自动同行竞争者的竞争中取得了有竞争力的分类准确性。此外，在所提出的算法在寻找最佳 CNN 架构方面比大多数同行竞争者消耗的计算资源要少得多。
