Using Uncertainty to Interpret your Model

This is a joint post with Inbar Naor. As deep neural networks (DNN) become more powerful, their complexity increases. This complexity introduces new challenges, including model interpretability. Interpretability is crucial in order to build models that are more robust and resistant to adversarial attacks. Moreover, designing a model for a new, not well researched domain is challenging and being able to interpret what the model is doing can help us in the process.