一种紧凑的DNN：逼近GoogLeNet-Level的分类精度和领域自适应
A Compact DNN: Approaching GoogLeNet-Level Accuracy of Classification and Domain Adaptation
摘要：
最近，基于网络架构设计的DNN模型压缩，例如，SqueezeNet，吸引了很多关注。与众所周知的模型相比，在这些非常紧凑的网络上没有观察到图像分类的准确性下降。然而，一个新出现的问题是，这些模型压缩技术是否会损害DNN的学习能力，而不是对单个数据集上的图像进行分类。我们的初步实验表明，这些压缩方法可能会降低领域自适应（DA）的能力，虽然分类性能保持。因此，本文提出了一种新的紧凑网络结构和无监督DA方法。DNN建立在一个新的基本模块Conv-M上，该模块提供了更多样化的特征提取器，而不会显着增加参数。我们的DA方法的统一框架将同时学习跨域的不变性，减少特征表示的发散，并适应标签预测。DNN拥有410万个参数，仅为AlexNet的6.7%或GoogLeNet的59%。实验表明，我们的DNN在分类和DA上都获得了GoogLeNet级别的准确性，我们的DA方法略优于以前的竞争对手。总而言之，我们基于DNN的DA策略在流行的Office-31数据集上的18个DA任务中的16个上实现了最先进的技术。
英文摘要：
Recently, DNN model compression based on network architecture design, e.g., SqueezeNet, attracted a lot attention. No accuracy drop on image classification is observed on these extremely compact networks, compared to well-known models. An emerging question, however, is whether these model compression techniques hurt DNN's learning ability other than classifying images on a single dataset. Our preliminary experiment shows that these compression methods could degrade domain adaptation (DA) ability, though the classification performance is preserved. Therefore, we propose a new compact network architecture and unsupervised DA method in this paper. The DNN is built on a new basic module Conv-M which provides more diverse feature extractors without significantly increasing parameters. The unified framework of our DA method will simultaneously learn invariance across domains, reduce divergence of feature representations, and adapt label prediction. Our DNN has 4.1M parameters, which is only 6.7% of AlexNet or 59% of GoogLeNet. Experiments show that our DNN obtains GoogLeNet-level accuracy both on classification and DA, and our DA method slightly outperforms previous competitive ones. Put all together, our DA strategy based on our DNN achieves state-of-the-art on sixteen of total eighteen DA tasks on popular Office-31 and Office-Caltech datasets.
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。
