Goto

Collaborating Authors

 measuring per-unit interpretability

Measuring Per-Unit Interpretability at Scale Without Humans

Neural Information Processing Systems

In today's era, whatever we can measure at scale, we can optimize. So far, measuring the interpretability of units in deep neural networks (DNNs) for computer vision still requires direct human evaluation and is not scalable. As a result, the inner workings of DNNs remain a mystery despite the remarkable progress we have seen in their applications. In this work, we introduce the first scalable method to measure the per-unit interpretability in vision DNNs. This method does not require any human evaluations, yet its prediction correlates well with existing human interpretability measurements.

  artificial intelligence, machine learning, measuring per-unit interpretability, (2 more...)
  Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (0.63)

Add feedback