Poster
On the (In)fidelity and Sensitivity of Explanations
Chih-Kuan Yeh · Cheng-Yu Hsieh · Arun Suggala · David Inouye · Pradeep Ravikumar

Thu Dec 12th 05:00 -- 07:00 PM @ East Exhibition Hall B + C #116

We consider objective evaluation measures of saliency explanations for complex black-box machine learning models. We propose simple robust variants of two notions that have been considered in recent literature: (in)fidelity, and sensitivity. We analyze optimal explanations with respect to both these measures, and while the optimal explanation for sensitivity is a vacuous constant explanation, the optimal explanation for infidelity is a novel combination of two popular explanation methods. By varying the perturbation distribution that defines infidelity, we obtain novel explanations by optimizing infidelity, which we show to out-perform existing explanations in both quantitative and qualitative measurements. Another salient question given these measures is how to modify any given explanation to have better values with respect to these measures. We propose a simple modification based on lowering sensitivity, and moreover show that when done appropriately, we could simultaneously improve both sensitivity as well as fidelity.

Author Information

Chih-Kuan Yeh (Carnegie Mellon University)
Cheng-Yu Hsieh (National Taiwan University)
Arun Suggala (Carnegie Mellon University)
David Inouye (Carnegie Mellon University)
Pradeep Ravikumar (Carnegie Mellon University)

More from the Same Authors