Net2Vec: Quantifying and Explaining how Concepts are Encoded by Filters in Deep Neural Networks
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition(2018)
摘要
In an effort to understand the meaning of the intermediate representations captured by deep networks, recent papers have tried to associate specific semantic concepts to individual neural network filter responses, where interesting correlations are often found, largely by focusing on extremal filter responses. In this paper, we show that this approach can favor easy-to-interpret cases that are not necessarily representative of the average behavior of a representation. A more realistic but harder-to-study hypothesis is that semantic representations are distributed, and thus filters must be studied in conjunction. In order to investigate this idea while enabling systematic visualization and quantification of multiple filter responses, we introduce the Net2Vec framework, in which semantic concepts are mapped to vectorial embeddings based on corresponding filter responses. By studying such embeddings, we are able to show that 1., in most cases, multiple filters are required to code for a concept, that 2., often filters are not concept specific and help encode multiple concepts, and that 3., compared to single filter activations, filter embeddings are able to better characterize the meaning of a representation and its relationship to other concepts.
更多查看译文
关键词
single filter activations,deep neural networks,intermediate representations,specific semantic concepts,individual neural network filter responses,extremal filter responses,average behavior,harder-to-study hypothesis,semantic representations,systematic visualization,multiple filter responses,Net2Vec framework
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络