Aspects of the present disclosure relate to machine learning techniques for continuous implementation and training of a machine learning system for identifying the natural language meaning of visual content. A computer vision model or other suitable machine learning model can predict whether a given descriptor is associated with the visual content. A set of such models can be used to determine whether particular ones of a set of descriptors are associated with the visual content, with the determined descriptors representing a meaning of the visual content. This meaning can be refined based on a multi-armed bandit tracking and analyzing interactions between the visual content and users associated with certain personas related to the determined descriptors.

The technology disclosed relates to a variant pathogenicity classifier. The variant pathogenicity classifier comprises memory and runtime logic. The memory stores (i) a reference amino acid sequence of a protein, (ii) an alternative amino acid sequence of the protein that contains a variant amino acid caused by a variant nucleotide, and (iii) a protein contact map of the protein. The runtime logic has access to the memory, and is configured to provide (i) the reference amino acid sequence, (ii) the alternative amino acid sequence, and (iii) the protein contact map as input to a first neural network, and to cause the first neural network to generate a pathogenicity indication of the variant amino acid as output in response to processing (i) the reference amino acid sequence, (ii) the alternative amino acid sequence, and (iii) the protein contact map.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for few-shot learning-based generator training based on raw data collected from a specific domain or class. In cases where the raw data is collected from multiple domains but is not easily divisible into classes, the invention describes training multiple generators based on a pivot-sample-based training process. Pivot samples are randomly selected from the raw data for clustering, and each cluster of raw data may be used to train a generator using the few-shot learning-based training process.

A device for machine learning is provided, including a first neural network layer, a second neural network layer with a normalization layer arranged in between. The normalization layer is configured to, when the device is undergoing training on a batch of training samples, receive multiple outputs of the first neural network layer for a plurality of training samples of the batch, each output comprising multiple data values for different indices on a first dimension and a second dimension; group the outputs into multiple groups based on the indices on the first and second dimensions; form a normalization output for each group which are provided as input to the second neural network layer. According to the application, the training of a deep convolutional neural network with good performance that performs stably at different batch sizes and is generalizable to multiple vision tasks is achieved, thereby improving the performance of the training.

An automated system for resolving program merges uses a multi-task neural transformer with attention. Each component of a merge conflict tuple (A, B, O) is represented as an AST and transformed into aligned AST-node sequences and aligned editing sequences. The multi-task neural transformer model predicts the tree editing steps needed to resolve the merge conflict and applies them to the AST representation of the code base. The tree editing steps include the edit actions that needed to be applied to the AST of the code base and the edit labels that are inserted or updated with the edit actions.

This disclosure relates to implementing a configuration portfolio having a compact set of model configurations that are predicted to perform well with respect to a wide variety of input tasks. Systems described herein involve evaluating machine learning models with respect to a set of training tasks to generate a regret matrix based on accuracy of the machine learning models in connection with predicting outputs for the training tasks. The systems described herein can identify a subset of model configurations from a plurality of model configurations based on the subset of model configurations having lower associated metrics of regret with respect to the training tasks. This ensures that each model configuration within the configuration portfolio will perform reasonably well for a given input task and provides a mechanism for selecting an output model configuration using significantly fewer processing resources than conventional model selection systems.

A learning processing device and method achieves learning of a lightweight model that is completed in a short amount of time. The learning processing device obtains a new, second learning model from an existing first learning model. An input unit acquires a first learning model generated in advance by learning a first learning data set, and an unpruned neural network (hereinafter, NN). An important parameter identification unit uses the first learning model and the NN to initialize a NN to be learned, and uses a second learning data set and the initialized NN to identify a degree of importance of parameters in a recognition process of the initialized NN. A new model generation unit carries out a pruning process for deleting parameters which are not important from the initialized NN, thereby generating a second NN; and a learning unit uses the second learning data set to learn the second NN.

A multi-task deep learning-based real-time matting method for non-green-screen portraits is provided. The method includes: performing binary classification adjustment on an original dataset, inputting an image or video containing portrait information, and performing preprocessing; constructing a deep learning network for person detection, extracting image features by using a deep residual neural network, and obtaining a region of interest (ROI) of portrait foreground and a portrait trimap in the ROI through logistic regression; and constructing a portrait alpha mask matting deep learning network. An encoder sharing mechanism effectively accelerates a computing process of the network. An alpha mask prediction result of the portrait foreground is output in an end-to-end manner to implement portrait matting. In this method, green screens are not required during portrait matting. In addition, during the matting, only original images or videos need to be provided, without a need to provide manually annotated portrait trimaps.

A computer-implemented method of machine-learning is described that includes obtaining a dataset of virtual scenes. The dataset of virtual scenes belongs to a first domain. The method further includes obtaining a test dataset of real scenes. The test dataset belongs to a second domain. The method further includes determining a third domain. The third domain is closer to the second domain than the first domain in terms of data distributions. The method further includes learning a domain-adaptive neural network based on the third domain. The domain-adaptive neural network is a neural network configured for inference of spatially reconfigurable objects in a real scene. Such a method constitutes an improved method of machine learning with a dataset of scenes including spatially reconfigurable objects.

Techniques are generally described for causal impact estimation using machine learning. A first machine learning model is trained using non-treatment variables during training. A second machine learning model uses learned weights from the first machine learning model for non-treatment variables and is trained on one or more treatment variables. The second machine learning model estimates outcomes based on the presence or absence of an event represented by the treatment variable. Selection bias is reduced by warm-starting the second machine learning model with non-treatment variable weights learned during training of the first machine learning model.