arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1802.01548cs.NEcs.AIcs.CVcs.DC
收藏

基于正则化进化的图像分类器结构搜索

Regularized Evolution for Image Classifier Architecture Search

Google Brain(谷歌大脑)

Esteban Real, Alok Aggarwal, Yanping Huang, Quoc V Le

摘要:

致力于手工制作神经网络图像分类器的努力,激发了使用架构搜索来自动发现它们。虽然进化算法已经被反复应用于神经网络拓扑结构,因此发现的图像分类器仍然不如人工制作的。在这里,我们进化出一个图像分类器-AmoebaNet-A-第一次超越了手工设计。为此,我们修改了比赛选择进化算法,通过引入年龄属性,有利于年轻的基因型。AmoebaNet-A的匹配大小与当前最先进的ImageNet模型具有相当的准确性,这些模型是通过更复杂的架构搜索方法发现的。扩展到更大的尺寸,AmoebaNet-A设置了最先进的83.9% / 96.6%的前5名ImageNet准确率。在与一个众所周知的强化学习算法的对照比较中,我们证明了进化可以用相同的硬件更快地获得结果,特别是在搜索的早期阶段。当可用的计算资源较少时,这是相关的。因此,进化是一种有效发现高质量架构的简单方法。

英文摘要:

The effort devoted to hand-crafting neural network image classifiers has motivated the use of architecture search to discover them automatically. Although evolutionary algorithms have been repeatedly applied to neural network topologies, the image classifiers thus discovered have remained inferior to human-crafted ones. Here, we evolve an image classifier---AmoebaNet-A---that surpasses hand-designs for the first time. To do this, we modify the tournament selection evolutionary algorithm by introducing an age property to favor the younger genotypes. Matching size, AmoebaNet-A has comparable accuracy to current state-of-the-art ImageNet models discovered with more complex architecture-search methods. Scaled to larger size, AmoebaNet-A sets a new state-of-the-art 83.9% / 96.6% top-5 ImageNet accuracy. In a controlled comparison against a well known reinforcement learning algorithm, we give evidence that evolution can obtain results faster with the same hardware, especially at the earlier stages of the search. This is relevant when fewer compute resources are available. Evolution is, thus, a simple method to effectively discover high-quality architectures.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑