arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.09824cs.LGcs.CV
收藏

自竞争神经网络

Self-Competitive Neural Networks

University of Tehran(德黑兰大学)

Iman Saberi, Fathiyeh Faghih

摘要:

深度神经网络(DNN)在许多应用中提高了分类问题的准确性。训练DNN的挑战之一是它需要由丰富的数据集提供,以提高其准确性并避免过度拟合。提高DNN泛化能力的一种方法是用新的合成对抗样本来增强训练数据。最近,研究人员已经广泛地提出了数据增强的方法。在本文中,我们生成对抗样本来细化每个类的吸引域(DoAs)。在这种方法中,在每个阶段,我们使用由主要和生成的对抗数据(直到该阶段)学习的模型,以DNN看起来复杂的方式操纵主要数据。然后,DNN使用增强的数据进行重新训练,然后它再次生成难以预测的对抗数据。由于DNN试图通过与自身竞争(生成硬样本,然后学习它们)来提高其准确性,因此该技术被称为自竞争神经网络(SCNN)。为了生成这样的样本,我们将问题作为一个优化任务,其中网络权重是固定的,并使用基于梯度下降的方法来合成处于其真实标签和最近错误标签边界上的对抗样本。我们的实验结果表明,使用SCNN的数据增强可以显着提高原始网络的精度。作为一个例子,我们可以提到将用MNIST数据集的1000个有限训练数据训练的CNN的准确度从94.26%提高到98.25%。

英文摘要:

Deep Neural Networks (DNNs) have improved the accuracy of classification problems in lots of applications. One of the challenges in training a DNN is its need to be fed by an enriched dataset to increase its accuracy and avoid it suffering from overfitting. One way to improve the generalization of DNNs is to augment the training data with new synthesized adversarial samples. Recently, researchers have worked extensively to propose methods for data augmentation. In this paper, we generate adversarial samples to refine the Domains of Attraction (DoAs) of each class. In this approach, at each stage, we use the model learned by the primary and generated adversarial data (up to that stage) to manipulate the primary data in a way that look complicated to the DNN. The DNN is then retrained using the augmented data and then it again generates adversarial data that are hard to predict for itself. As the DNN tries to improve its accuracy by competing with itself (generating hard samples and then learning them), the technique is called Self-Competitive Neural Network (SCNN). To generate such samples, we pose the problem as an optimization task, where the network weights are fixed and use a gradient descent based method to synthesize adversarial samples that are on the boundary of their true labels and the nearest wrong labels. Our experimental results show that data augmentation using SCNNs can significantly increase the accuracy of the original network. As an example, we can mention improving the accuracy of a CNN trained with 1000 limited training data of MNIST dataset from 94.26% to 98.25%.

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。

↑