Convolutional Neural Networks have been a subject of great importance over
the past decade and great strides have been made in their utility for producing
state of the art performance in many computer vision problems. However, the
behavior of deep networks is yet to be fully understood and is still an active
area of research. In this work, we present an intriguing behavior: pre-trained
CNNs can be made to improve their predictions by structurally perturbing the
input. We observe that these perturbations - referred as Guided Perturbations -
enable a trained network to improve its prediction performance without any
learning or change in network weights. We perform various ablative experiments
to understand how these perturbations affect the local context and feature
representations. Furthermore, we demonstrate that this idea can improve
performance of several existing approaches on semantic segmentation and scene
labeling tasks on the PASCAL VOC dataset and supervised classification tasks on
MNIST and CIFAR10 datasets.

本篇论文探究了卷积神经网络中的一种新行为：通过结构化扰动输入，预训练的 CNN 可以改善其预测性能，这种扰动被称为 Guided Perturbations。实验结果表明，这种方法可以提高现有方法在语义分割和场景标注任务上的性能，并且可以应用于 MNIST 和 CIFAR10 等数据集上的分类任务。