arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.08446cs.CV
收藏

基于滤波器分布模板的高效卷积网络模型

Towards Efficient Convolutional Network Models with Filter Distribution Templates

University of Bristol(布里斯托大学)

Ramon Izquierdo-Cordova, Walterio Mayol-Cuevas

摘要:

当特征图减少时,增加更深层的过滤器数量是卷积网络设计中广泛采用的模式。它可以在经典的CNN架构和自动发现模型中找到。即使CNS方法通常也会探索来自这种金字塔模式的乘数的选择。我们通过引入一小组模板来对抗这种做法,这些模板由VGG和ResNet架构中过滤器原始金字塔分布的易于实现,直观和积极的变化组成。在CIFAR,CINIC10和TinyImagenet数据集上的实验表明,由我们的模板生成的模型在更少的参数和内存需求方面更有效。

英文摘要:

Increasing number of filters in deeper layers when feature maps are decreased is a widely adopted pattern in convolutional network design. It can be found in classical CNN architectures and in automatic discovered models. Even CNS methods commonly explore a selection of multipliers derived from this pyramidal pattern. We defy this practice by introducing a small set of templates consisting of easy to implement, intuitive and aggressive variations of the original pyramidal distribution of filters in VGG and ResNet architectures. Experiments on CIFAR, CINIC10 and TinyImagenet datasets show that models produced by our templates, are more efficient in terms of fewer parameters and memory needs.

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。

↑