"A Benchmark for Interpretability Methods in Deep Neural Networks."

Sara Hooker et al. (2019)
a service of Schloss Dagstuhl - Leibniz Center for Informatics