Figure (IMAGE) U.S. Army Research Laboratory Caption In a robust tomography scheme with machine learning, noisy tomography measurements are fed to the convolutional neural network, which makes predictions of intermediate t-matrices as the outputs. At the end, the predicted matrices are inverted to reconstruct the pure density matrices for the given noisy measurements. Credit U.S. Army image Usage Restrictions None License Licensed content Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.