Proceedings of the 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016)