most citedImproving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization

59 citations · 59 across the 1 of their papers we have counts for

collaborators

1 paper

eess.IV2022★ 59 cited

Improving deep neural network generalization and robustness to background bias via layer-wise relevance propagation optimization

Pedro R. A. S. Bassi, Sergio S. J. Dertkigil, Andrea Cavalli

Features in images' backgrounds can spuriously correlate with the images' classes, representing background bias. They can influence the classifier's decisions, causing shortcut lea…