Exploring RNN Interpretability Methods Proposed by Zhou Zhihua et al.
Selected from ArXiv Authors: Bo-Jian Hou, Zhi-Hua Zhou Contributors: Si Yuan, Xiao Kun This article is authorized for reproduction by Almost Human (almosthuman2014) Reproduction is prohibited Apart from numerical calculations, do you really know what neural networks are doing internally? We have always understood deep models based on their computational flow, but we are still … Read more