Abstract: Methods and systems for classifying a long-tail distribution of data. Data deriving from one or more sensors is classified into a plurality of classes by using (i) a feature-extractor backbone model configured to extract features from the data, and (ii) a classifier model configured to classify the data based on the extracted features. The plurality of classes are grouped, with each group assigned to a respective teacher model. Each respective teacher model is trained with the data in its respective group, as well as the feature-extractor backbone model. The outputs of the teacher models are then merged into a final class prediction model configured to classify the data.

Abstract: Self-supervised training of machine learning (“ML”) algorithms for reconstruction in inverse problems are described. These techniques do not require fully sampled training data. As an example, a physics-based ML reconstruction can be trained without requiring fully-sampled training data. In this way, such ML-based reconstruction algorithms can be trained on existing databases of undersampled images or in a scan-specific manner.