Search Results for author:
Found 3 papers, 3 papers with code
Attri-Net: A Globally and Locally Inherently Interpretable Model for Multi-Label Classification Using Class-Specific Counterfactuals
Attri-Net first counterfactually generates class-specific attribution maps to highlight the disease evidence, then performs classification with logistic regression classifiers based solely on the attribution maps.
Right for the Wrong Reason: Can Interpretable ML Techniques Detect Spurious Correlations?
Such dependencies on confounding information can be difficult to detect using performance metrics if the test data comes from the same distribution as the training data.
Inherently Interpretable Multi-Label Classification Using Class-Specific Counterfactuals
Furthermore, as we show in this paper, current explanation techniques do not perform adequately in the multi-label scenario, in which multiple medical findings may co-occur in a single image.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
