With the advent of highly predictive but opaque deep learning models, it has become more important than ever to understand and explain the predictions of such models. Existing approaches define interpretability as the inverse of complexity and achieve interpretability at the cost of accuracy. This introduces a risk of producing interpretable but misleading explanations. As humans, we are prone to engage in this kind of behavior \cite{mythos}. In this paper, we take a step in the direction of tackling the problem of interpretability without compromising the model accuracy. We propose to build a Treeview representation of the complex model via hierarchical partitioning of the feature space, which reveals the iterative rejection of unlikely class labels until the correct association is predicted.
[1]
Cynthia Rudin,et al.
Methods and Models for Interpretable Linear Classification
,
2014,
ArXiv.
[2]
Kush R. Varshney,et al.
Exact Rule Learning via Boolean Compressed Sensing
,
2013,
ICML.
[3]
Carlos Guestrin,et al.
Model-Agnostic Interpretability of Machine Learning
,
2016,
ArXiv.
[4]
Zachary Chase Lipton.
The mythos of model interpretability
,
2016,
ACM Queue.
[5]
Ronald L. Rivest,et al.
Learning decision lists
,
2004,
Machine Learning.