← Search

Elizabeth Purdom

2 accepted papers

2026

When Random Saliency Looks Trained: Architectural Center Bias in CNN Interpretability

ICML 2026poster

Saliency maps are widely used to interpret image classification models and build trust in their predictions; however, their reliability remains a central concern, as randomized networks can produce saliency maps that closely resemble those of trained models. We identify a previously underappreciated…

Cited by 0SourceScholar
2019

The non-parametric bootstrap and spectral analysis in moderate and high-dimension

AISTATS 2019poster

We consider the properties of the bootstrap as a tool for inference concerning the eigenvalues of a sample covariance matrix computed from an n x p data matrix X. We focus on the modern framework where p/n is not close to 0 but remains bounded as n and p tend to infinity. Through a mix of numerical…

Cited by 13SourcePDFScholar