4.8 Article

Building and Interpreting Deep Similarity Models

出版社

IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2020.3020738

关键词

Machine learning; Data models; Robustness; Neural networks; Taylor series; Feature extraction; Deep learning; Similarity; layer-wise relevance propagation; deep neural networks; explainable machine learning; digital humanities

资金

  1. German Ministry for Education and Research [01IS18025A, 01IS18037A]
  2. German Research Foundation (DFG) as Math+: Berlin Mathematics Research Center [390685689]
  3. Institute for Information & Communications Technology Planning & Evaluation (IITP) - Korea Government [2019-0-00079]

向作者/读者索取更多资源

This paper proposes a method to make similarities interpretable by decomposing deep similarity models and provides insights into complex similarity models. The method is applied to assess similarity between historical documents in digital humanities.
Many learning algorithms such as kernel machines, nearest neighbors, clustering, or anomaly detection, are based on distances or similarities. Before similarities are used for training an actual machine learning model, we would like to verify that they are bound to meaningful patterns in the data. In this paper, we propose to make similarities interpretable by augmenting them with an explanation. We develop BiLRP, a scalable and theoretically founded method to systematically decompose the output of an already trained deep similarity model on pairs of input features. Our method can be expressed as a composition of LRP explanations, which were shown in previous works to scale to highly nonlinear models. Through an extensive set of experiments, we demonstrate that BiLRP robustly explains complex similarity models, e.g., built on VGG-16 deep neural network features. Additionally, we apply our method to an open problem in digital humanities: detailed assessment of similarity between historical documents, such as astronomical tables. Here again, BiLRP provides insight and brings verifiability into a highly engineered and problem-specific similarity model.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.8
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据