arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1709.05360cs.CVcs.AI
收藏

将深度网络嵌入视觉系统

Embedding Deep Networks into Visual Explanations

Oregon State University(俄勒冈州立大学);Tencent(腾讯)

Zhongang Qi, Saeed Khorram, Fuxin Li

摘要:

在本文中,我们提出了一种新的解释神经网络(XNN)来解释深度网络所做的预测。XNN通过学习将深度网络层的高维激活向量非线性嵌入到低维解释空间中来工作,同时保持忠实性,即,原始的深度学习预测可以从我们的解释网络提取的几个概念中构建。然后,我们将这些概念可视化,让人类了解深层网络用于决策的高级概念。我们提出了一个算法称为稀疏重建自动编码器(SRAE)学习嵌入的解释空间。SRAE的目标是重建原始特征空间的一部分,同时保持忠实性。在SRAE中引入拉离项,使解释空间的基更加正交。一个可视化系统,然后介绍了人类的理解的解释空间中的功能。所提出的方法被应用于解释图像分类任务中的CNN模型。我们进行了一项人体研究,结果表明,该方法优于单一显着性图基线,并提高了人类在困难的分类任务上的表现。此外,几个新的指标被引入到定量评估的解释性能没有人的参与。

英文摘要:

In this paper, we propose a novel Explanation Neural Network (XNN) to explain the predictions made by a deep network. The XNN works by learning a nonlinear embedding of a high-dimensional activation vector of a deep network layer into a low-dimensional explanation space while retaining faithfulness i.e., the original deep learning predictions can be constructed from the few concepts extracted by our explanation network. We then visualize such concepts for human to learn about the high-level concepts that the deep network is using to make decisions. We propose an algorithm called Sparse Reconstruction Autoencoder (SRAE) for learning the embedding to the explanation space. SRAE aims to reconstruct part of the original feature space while retaining faithfulness. A pull-away term is applied to SRAE to make the bases of the explanation space more orthogonal to each other. A visualization system is then introduced for human understanding of the features in the explanation space. The proposed method is applied to explain CNN models in image classification tasks. We conducted a human study, which shows that the proposed approach outperforms single saliency map baselines, and improves human performance on a difficult classification tasks. Also, several novel metrics are introduced to evaluate the performance of explanations quantitatively without human involvement.

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。

↑