arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1312.5355cs.NEcs.CV
收藏

用于深度学习的生成式NeuroEvolution

Generative NeuroEvolution for Deep Learning

Space and Naval Warfare Systems Center – Pacific(太平洋空间与海战系统中心)

Phillip Verbancsics, Josh Harguess

摘要:

机器学习(ML)社区的一个重要目标是创建可以学习具有人类水平能力的解决方案的方法。人类拥有显著优势的一个领域是视觉处理。解决这一差距的一个重要方法是机器学习方法,这些方法受到自然系统的启发,例如人工神经网络(ANN),演化计算(EC)以及生成和发展系统(GDS)。对深度学习的研究表明,这种架构可以在某些视觉任务上实现与人类竞争的性能;然而,这些系统主要是通过监督和无监督学习算法进行训练的。另一方面,研究表明,进化可能在视觉系统的发展中起着重要作用。因此,本文研究了神经进化(NE)在深度学习中的作用。特别地,基于超立方体的增强拓扑的神经进化是一种NE方法,其可以通过训练将ANN权重模式压缩为几何形状的函数的间接编码来有效地学习大型神经结构。结果表明,HyperNEAT很难单独执行图像分类,但可以有效地训练其他ML方法可以学习的特征提取器。因此,NeuroEvolution与其他ML方法相结合提供了一个有趣的研究领域,可以复制自然界中的过程。

英文摘要:

An important goal for the machine learning (ML) community is to create approaches that can learn solutions with human-level capability. One domain where humans have held a significant advantage is visual processing. A significant approach to addressing this gap has been machine learning approaches that are inspired from the natural systems, such as artificial neural networks (ANNs), evolutionary computation (EC), and generative and developmental systems (GDS). Research into deep learning has demonstrated that such architectures can achieve performance competitive with humans on some visual tasks; however, these systems have been primarily trained through supervised and unsupervised learning algorithms. Alternatively, research is showing that evolution may have a significant role in the development of visual systems. Thus this paper investigates the role neuro-evolution (NE) can take in deep learning. In particular, the Hypercube-based NeuroEvolution of Augmenting Topologies is a NE approach that can effectively learn large neural structures by training an indirect encoding that compresses the ANN weight pattern as a function of geometry. The results show that HyperNEAT struggles with performing image classification by itself, but can be effective in training a feature extractor that other ML approaches can learn from. Thus NeuroEvolution combined with other ML methods provides an intriguing area of research that can replicate the processes in nature.

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。

↑