A Benchmark For Interpretability Methods In Deep Neural Networks
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019)(2019)
摘要
We propose an empirical measure of the approximate accuracy of feature importance estimates in deep neural networks. Our results across several large-scale image classification datasets show that many popular interpretability methods produce estimates of feature importance that are not better than a random designation of feature importance. Only certain ensemble based approaches-VarGrad and SmoothGrad-Squared-outperform such a random assignment of importance. The manner of ensembling remains critical, we show that some approaches do no better then the underlying method but carry a far higher computational burden.
更多查看译文
关键词
random assignment
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络