InterpNET：用于可解释深度学习的神经内省
InterpNET: Neural Introspection for Interpretable Deep Learning
摘要：
人类能够解释他们的推理。相反，深度神经网络不是。本文试图通过引入一种新的方法来设计可解释的神经网络分类，人类视觉系统的内部工作的生理证据的启发，以弥合这一差距。本文提出了一种神经网络设计范式，称为InterpNET，它可以与任何现有的分类体系结构相结合，产生自然语言的分类解释。该模块的成功依赖于网络的计算和推理在其内部层激活中表示的假设。虽然原则上InterpNET可以应用于任何现有的分类体系结构，它是通过一个图像分类和解释任务进行评估。在CUB鸟类分类和解释数据集上的实验表明，该模型能够定性和定量地生成高质量的解释。虽然目前该数据集的最新METEOR评分为29.2，但InterpNET的METEOR评分要高得多，为37.9。
英文摘要：
Humans are able to explain their reasoning. On the contrary, deep neural networks are not. This paper attempts to bridge this gap by introducing a new way to design interpretable neural networks for classification, inspired by physiological evidence of the human visual system's inner-workings. This paper proposes a neural network design paradigm, termed InterpNET, which can be combined with any existing classification architecture to generate natural language explanations of the classifications. The success of the module relies on the assumption that the network's computation and reasoning is represented in its internal layer activations. While in principle InterpNET could be applied to any existing classification architecture, it is evaluated via an image classification and explanation task. Experiments on a CUB bird classification and explanation dataset show qualitatively and quantitatively that the model is able to generate high-quality explanations. While the current state-of-the-art METEOR score on this dataset is 29.2, InterpNET achieves a much higher METEOR score of 37.9.
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。
