当前位置: X-MOL 学术Pattern Recogn. Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Solving the same-different task with convolutional neural networks
Pattern Recognition Letters ( IF 3.9 ) Pub Date : 2021-01-07 , DOI: 10.1016/j.patrec.2020.12.019
Nicola Messina , Giuseppe Amato , Fabio Carrara , Claudio Gennaro , Fabrizio Falchi

Deep learning demonstrated major abilities in solving many kinds of different real-world problems in computer vision literature. However, they are still strained by simple reasoning tasks that humans consider easy to solve. In this work, we probe current state-of-the-art convolutional neural networks on a difficult set of tasks known as the same-different problems. All the problems require the same prerequisite to be solved correctly: understanding if two random shapes inside the same image are the same or not. With the experiments carried out in this work, we demonstrate that residual connections, and more generally the skip connections, seem to have only a marginal impact on the learning of the proposed problems. In particular, we experiment with DenseNets, and we examine the contribution of residual and recurrent connections in already tested architectures, ResNet-18, and CorNet-S respectively. Our experiments show that older feed-forward networks, AlexNet and VGG, are almost unable to learn the proposed problems, except in some specific scenarios. We show that recently introduced architectures can converge even in the cases where the important parts of their architecture are removed. We finally carry out some zero-shot generalization tests, and we discover that in these scenarios residual and recurrent connections can have a stronger impact on the overall test accuracy. On four difficult problems from the SVRT dataset, we can reach state-of-the-art results with respect to the previous approaches, obtaining super-human performances on three of the four problems.



中文翻译:

用卷积神经网络解决同一个任务

深度学习证明了解决计算机视觉文学中许多不同现实问题的主要能力。但是,它们仍然受人类认为易于解决的简单推理任务的困扰。在这项工作中,我们研究了一系列困难的任务,这些任务被称为同差法,从而研究了当前最新的卷积神经网络。问题。所有问题都需要相同的先决条件才能正确解决:了解同一图像内的两个随机形状是否相同。通过这项工作中进行的实验,我们证明了残留连接,更普遍地说是跳过连接,似乎对所提出问题的学习仅具有边际影响。特别是,我们对DenseNets进行了试验,并分别检查了已测试的架构ResNet-18和CorNet-S中的残余连接和循环连接。我们的实验表明,除某些特定情况外,较旧的前馈网络AlexNet和VGG几乎无法了解所提出的问题。我们证明,即使删除了架构的重要部分,最近引入的架构也可以收敛。最后,我们进行了一些零散泛化测试,我们发现在这些情况下,残余连接和循环连接对整体测试精度有更强的影响。在SVRT数据集中的四个困难问题上,我们可以获得相对于先前方法的最新结果,从而在四个问题中的三个问题上获得了超人的表现。

更新日期:2021-01-20
down
wechat
bug