Resources on ML interpretability

This section serves the purpose of gathering recent resources on interpretability in machine learning to stimulate further research on this important emerging topic.

• Recently, Pereira and Meier approached the interpretability of ML on the problems of Brain Tumor and Stroke segmentation.
• The excellent article of Patrick Hall (https://www.h2o.ai/) on the wide range of different approaches to tackle interpretability of machine learning models.
• Last year’s ICML workshop on Human Interpretability in machine learning (WHI) featuring online workshop proceedings.
• Website of the annual event on Fairness, Accountability, and Transparency in Machine Learning.
• Website of joint research project at Frauenhofer HHI, TU Berlin, and SUTD Singapore on better understanding predictions of machine learning models.
• Last year’s interpretable machine learning symposium at NIPS conference, including workshop papers and a debate on the importance of interpretability.
• Also, the first NIPS debate on the necessity of interpetability for machine learning.