Can I read Deep Learning Robustness in Inverse Problems on EtoBox?

Deep Learning Robustness in Inverse Problems by Subhosri Basu is a document available to read on EtoBox.

What is Deep Learning Robustness in Inverse Problems about?

The document discusses the robustness of reconstruction algorithms in linear inverse problems, focusing on the differences between worst-case and average-case perturbations. It highlights that worst-case perturbations can significantly affect reconstruction accuracy, while average-case perturbations tend to have a smaller impact. The findings emphasize the importance of the measurement matrix and its interaction with the reconstruction algorithm in determining robustness.

Author

Subhosri Basu

Language

EN