Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a machine learning model to perform a machine learning task by processing input data to the model. For example, the input data can include image, video, or point cloud data, and the task can be a perception task such as classification or detection task. In one aspect, the method includes receiving training data including a plurality of training inputs; receiving a plurality of data augmentation policy parameters that define different transformation operations for transforming training inputs before the training inputs are used to train the machine learning model; maintaining a plurality of candidate machine learning models; for each of the plurality of candidate machine learning models: repeatedly determining an augmented batch of training data; training the candidate machine learning model using the augmented batch of the training data; and updating the maintained data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating object interaction predictions using a neural network. One of the methods includes obtaining a sensor input derived from data generated by one or more sensors that characterizes a scene. The sensor input is provided to an object interaction neural network. The object interaction neural network is configured to process the sensor input to generate a plurality of object interaction outputs. Each respective object interaction output includes main object information and interacting object information. The respective object interaction outputs corresponding to the plurality of regions in the sensor input are received as output of the object interaction neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing data generated by a sensing system that rotationally senses an environment. In one aspect, a method comprises partitioning a predetermined period of time into a plurality of sub-periods, wherein the predetermined period of time is a period of time for which data generated by the sensing system constitutes a complete rotational sensing of the environment; for each sub-period: receiving current data generated by the sensing system during the sub-period and characterizing a respective partial scene of the environment; processing the current data using an object detection neural network to generate a current object detection output that is specific to the respective partial scene of the environment.

Abstract: A method for determining an architecture for a task neural network configured to perform a particular machine learning task is described.

Abstract: Systems and methods for transforming grayscale images into color images using deep neural networks are described. One of the systems include one or more computers and one or more storage devices storing instructions that, when executed by one or more computers, cause the one or more computers to implement a coloring neural network, a refinement neural network, and a subsystem. The coloring neural network is configured to receive a first grayscale image having a first resolution and to process the first grayscale image to generate a first color image having a second resolution lower than the first resolution. The subsystem processes the first color image to generate a set of intermediate image outputs. The refinement neural network is configured to receive the set intermediate image outputs, and to process the set of intermediate image outputs to generate a second color image having a third resolution higher than the second resolution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating object interaction predictions using a neural network. One of the methods includes obtaining a sensor input derived from data generated by one or more sensors that characterizes a scene. The sensor input is provided to an object interaction neural network. The object interaction neural network is configured to process the sensor input to generate a plurality of object interaction outputs. Each respective object interaction output includes main object information and interacting object information. The respective object interaction outputs corresponding to the plurality of regions in the sensor input are received as output of the object interaction neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating object interaction predictions using a neural network. One of the methods includes obtaining a sensor input derived from data generated by one or more sensors that characterizes a scene. The sensor input is provided to an object interaction neural network. The object interaction neural network is configured to process the sensor input to generate a plurality of object interaction outputs. Each respective object interaction output includes main object information and interacting object information. The respective object interaction outputs corresponding to the plurality of regions in the sensor input are received as output of the object interaction neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a machine learning model. One of the methods includes obtaining a training data set for training a machine learning model, the training data set comprising a plurality of training inputs; determining a plurality of data augmentation policies, wherein each data augmentation policy defines a procedure for processing a training input to generate a transformed training input; for each data augmentation policy, training the machine learning model using the data augmentation policy; determining, for each data augmentation policy, a quality measure of the machine learning model that has been trained using the data augmentation policy; and selecting a final data augmentation policy based using the quality measures of the machine learning models.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining neural network architectures. One of the methods includes obtaining training data for a dense image prediction task; and determining an architecture for a neural network configured to perform the dense image prediction task, comprising: searching a space of candidate architectures to identify one or more best performing architectures using the training data, wherein each candidate architecture in the space of candidate architectures comprises (i) the same first neural network backbone that is configured to receive an input image and to process the input image to generate a plurality of feature maps and (ii) a different dense prediction cell configured to process the plurality of feature maps and to generate an output for the dense image prediction task; and determining the architecture for the neural network based on the best performing candidate architectures.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing point cloud data representing a sensor measurement of a scene captured by one or more sensors to generate an object detection output that identifies locations of one or more objects in the scene. When deployed within an on-board system of a vehicle, the object detection output that is generated can be used to make autonomous driving decisions for the vehicle with enhanced accuracy.

Abstract: The specification describes methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a larger neural network from a smaller neural network. One of the described methods includes obtaining data specifying an original neural network and generating a larger neural network from the original neural network. The larger neural network has a larger neural network structure than the original neural network structure. The values of the parameters of the original neural network units and the additional neural network units are initialized so that the larger neural network generates the same outputs from the same inputs as the original neural network, and the larger neural network is trained to determine trained values of the parameters of the original neural network units and the additional neural network units from the initialized values.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying data objects. One of the methods includes obtaining data that associates each term in a vocabulary of terms with a respective high-dimensional representation of the term; obtaining classification data for a data object, wherein the classification data includes a respective score for each of a plurality of categories, and wherein each of the categories is associated with a respective category label; computing an aggregate high-dimensional representation for the data object from high-dimensional representations for the category labels associated with the categories and the respective scores; identifying a first term in the vocabulary of terms having a high-dimensional representation that is closest to the aggregate high-dimensional representation; and selecting the first term as a category label for the data object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining neural network architectures. One of the methods includes obtaining training data for a dense image prediction task; and determining an architecture for a neural network configured to perform the dense image prediction task, comprising: searching a space of candidate architectures to identify one or more best performing architectures using the training data, wherein each candidate architecture in the space of candidate architectures comprises (i) the same first neural network backbone that is configured to receive an input image and to process the input image to generate a plurality of feature maps and (ii) a different dense prediction cell configured to process the plurality of feature maps and to generate an output for the dense image prediction task; and determining the architecture for the neural network based on the best performing candidate architectures.

Abstract: A method for generating a segmentation of an image that assigns each pixel to a respective segmentation category from In a set of segmentation categories is described. The method includes obtaining features of the image, the image including a plurality of pixels. For each of one or more time steps starting from an initial time step and continuing until a final time step, the method includes generating a network input from the features of the image and a current segmentation output as of the time step, processing the network input using a convolutional recurrent neural network to generate an intermediate segmentation output for the time step, and generating an up dated segmentation output for the time step from the intermediate segmentation output for the time step and the current segmentation output as of the time step. The method includes generating a final segmentation of the image from the updated segmentation output.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying data objects. One of the methods includes obtaining data that associates each term in a vocabulary of terms with a respective high-dimensional representation of the term; obtaining classification data for a data object, wherein the classification data includes a respective score for each of a plurality of categories, and wherein each of the categories is associated with a respective category label; computing an aggregate high-dimensional representation for the data object from high-dimensional representations for the category labels associated with the categories and the respective scores; identifying a first term in the vocabulary of terms having a high-dimensional representation that is closest to the aggregate high-dimensional representation; and selecting the first term as a category label for the data object.

Abstract: A method for determining an architecture for a task neural network configured to perform a particular machine learning task is described.

Abstract: A method for applying a style to an input image to generate a stylized image. The method includes maintaining data specifying respective parameter values for each image style in a set of image styles, receiving an input including an input image and data identifying an input style to be applied to the input image to generate a stylized image that is in the input style, determining, from the maintained data, parameter values for the input style, and generating the stylized image by processing the input image using a style transfer neural network that is configured to process the input image to generate the stylized image.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining neural network architectures. One of the methods includes generating, using a controller neural network having controller parameters and in accordance with current values of the controller parameters, a batch of output sequences. The method includes, for each output sequence in the batch: generating an instance of a child convolutional neural network (CNN) that includes multiple instances of a first convolutional cell having an architecture defined by the output sequence; training the instance of the child CNN to perform an image processing task; and evaluating a performance of the trained instance of the child CNN on the task to determine a performance metric for the trained instance of the child CNN; and using the performance metrics for the trained instances of the child CNN to adjust current values of the controller parameters of the controller neural network.

Abstract: A method for applying a style to an input image to generate a stylized image. The method includes maintaining data specifying respective parameter values for each image style in a set of image styles, receiving an input including an input image and data identifying an input style to be applied to the input image to generate a stylized image that is in the input style, determining, from the maintained data, parameter values for the input style, and generating the stylized image by processing the input image using a style transfer neural network that is configured to process the input image to generate the stylized image.

Abstract: Systems and techniques are disclosed for labeling objects within an image. The objects may be labeled by selecting an option from a plurality of options such that each option is a potential label for the object. An option may have an option score associated with. Additionally, a relation score may be calculated for a first option and a second option corresponding to a second object in an image. The relation score may be based on a frequency, probability, or observance corresponding to the co-occurrence of text associated with the first option and the second option in a text corpus such as the World Wide Web. An option may be selected as a label for an object based on a global score calculated based at least on an option score and relation score associated with the option.