Abstract: In particular embodiments, a computing system may access a plurality of images for pre-training a first machine-learning model that includes an encoder and a decoder. Using each image, the system may pre-train the model by dividing the image into a set a patches, selecting a first subset of the patches to be visible and a second subset of the patches to be masked during the pre-training, processing, using the encoder, the first subset of patches to generate corresponding first latent representations, processing, using the decoder, the first latent representations corresponding to the first subset of patches and mask tokens corresponding to the second subset of patches to generate reconstructed patches corresponding to the second subset of patches, the reconstructed patches and the first subset of patches being used to generate a reconstructed image, and updating the model based on comparisons between the image and the reconstructed image.

Abstract: Techniques are described to generate improved training data for pixel labeling. To generate training data, objects are displayed in a user interface by a computing device, e.g., iteratively. The objects are taken from a structured object representation associated with a respective one of a plurality of images. The structured object representation defines a hierarchical relationship of the objects within the respective image. Inputs are then received that are originated through user interaction with the user interface. The inputs label respective ones of the iteratively displayed objects, e.g., as text, a graphical element, background, foreground, and so forth. A model is trained by the computing device using machine learning.

Abstract: Techniques are described to generate improved training data for pixel labeling. To generate training data, objects are displayed in a user interface by a computing device, e.g., iteratively. The objects are taken from a structured object representation associated with a respective one of a plurality of images. The structured object representation defines a hierarchical relationship of the objects within the respective image. Inputs are then received that are originated through user interaction with the user interface. The inputs label respective ones of the iteratively displayed objects, e.g., as text, a graphical element, background, foreground, and so forth. A model is trained by the computing device using machine learning.

Abstract: Systems and methods are disclosed for training a learning machine by augmenting data from fine-grained image recognition with labeled data annotated by one or more hyper-classes, performing multi-task deep learning; allowing fine-grained classification and hyper-class classification to share and learn the same feature layers; and applying regularization in the multi-task deep learning to exploit one or more relationships between the fine-grained classes and the hyper-classes.