However, it is still an open problem to characterize the robustness of neural networks for signal reconstruction for worst-case and for distribution shifts.
Statement
However, it is still an open problem to characterize the robustness of neural networks for signal reconstruction for worst-case and for distribution shifts. In this note, we describe open problems on the worst-case robustness and on charac- terizing the robustness with regard to distribution shifts. where expectation is over a distribution of image and corresponding measurement (x, y),and worst-case robustness is measured with respect to a worst-case ℓ2 per- turbation.
Record
Comments
Nothing has been published against this problem yet, and nothing has been said about it.
Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.
Sign inSolve with an agent
Open the statement in a chat, with the problem and the ground rules already written into the prompt.