Rare event training data sets for robust training of semiconductor yield related components

A non-transitory computer-readable medium, storing program instructions executable on one or more computer systems for performing a computer-implemented method for determining information for a specimen, wherein the computer-implemented method comprises:

• collecting images of known rare defect types previously detected on one or more other specimens;
• assigning training labels to the images responsive to the known rare defect types in the images;
• and
• storing the collected images and the assigned training labels as a rare defect type training data set, wherein the rare defect type training data set is unsuitable for use in training a component configured for determining information for the specimen from runtime images generated for the specimen by an imaging system until the rare defect type training data set is combined with training images and corresponding training labels generated for the specimen.

A computer-implemented method for determining information for a specimen, comprising:

• collecting images of known rare defect types previously detected on one or more other specimens;
• assigning training labels to the images responsive to the known rare defect types in the images;
• and
• storing the collected images and the assigned training labels as a rare defect type training data set, wherein the rare defect type training data set is unsuitable for use in training a component configured for determining information for the specimen from runtime images generated for the specimen by an imaging system until the rare defect type training data set is combined with training images and corresponding training labels generated for the specimen, and wherein said collecting, assigning, and storing are performed by one or more computer systems.