Abstract: A helmet selection/customization process, which comprises obtaining a 3D scan of a user's head; identifying a retail helmet variant that would fit the user's head; providing the user with an option to select the identified retail helmet variant or design a custom liner for a base helmet; and in case the user selects to design a custom helmet, causing generation of a 3D model of a custom liner based on the 3D scan and a 3D model of the base helmet. The method includes determining, based on the 3D scan, parameters associated with the user's head; and accessing a database storing parameters associated with a plurality of retail helmet variants, wherein the identifying is carried out based on processing of the parameters associated with the user's head and the parameters stored in the database, so as to identify one of the variants in the plurality of retail helmet variants.

Abstract: Methods and systems for performing a sequence of machine learning tasks. One system includes a sequence of deep neural networks (DNNs), including: a first DNN corresponding to a first machine learning task, wherein the first DNN comprises a first plurality of indexed layers, and each layer in the first plurality of indexed layers is configured to receive a respective layer input and process the layer input to generate a respective layer output; and one or more subsequent DNNs corresponding to one or more respective machine learning tasks, wherein each subsequent DNN comprises a respective plurality of indexed layers, and each layer in a respective plurality of indexed layers with index greater than one receives input from a preceding layer of the respective subsequent DNN, and one or more preceding layers of respective preceding DNNs, wherein a preceding layer is a layer whose index is one less than the current index.

Abstract: Methods and systems for performing a sequence of machine learning tasks. One system includes a sequence of deep neural networks (DNNs), including: a first DNN corresponding to a first machine learning task, wherein the first DNN comprises a first plurality of indexed layers, and each layer in the first plurality of indexed layers is configured to receive a respective layer input and process the layer input to generate a respective layer output; and one or more subsequent DNNs corresponding to one or more respective machine learning tasks, wherein each subsequent DNN comprises a respective plurality of indexed layers, and each layer in a respective plurality of indexed layers with index greater than one receives input from a preceding layer of the respective subsequent DNN, and one or more preceding layers of respective preceding DNNs, wherein a preceding layer is a layer whose index is one less than the current index.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes one or more transformed activation function layers.

Abstract: Methods and systems for performing a sequence of machine learning tasks. One system includes a sequence of deep neural networks (DNNs), including: a first DNN corresponding to a first machine learning task, wherein the first DNN comprises a first plurality of indexed layers, and each layer in the first plurality of indexed layers is configured to receive a respective layer input and process the layer input to generate a respective layer output; and one or more subsequent DNNs corresponding to one or more respective machine learning tasks, wherein each subsequent DNN comprises a respective plurality of indexed layers, and each layer in a respective plurality of indexed layers with index greater than one receives input from a preceding layer of the respective subsequent DNN, and one or more preceding layers of respective preceding DNNs, wherein a preceding layer is a layer whose index is one less than the current index.

Abstract: Methods and systems for performing a sequence of machine learning tasks. One system includes a sequence of deep neural networks (DNNs), including: a first DNN corresponding to a first machine learning task, wherein the first DNN comprises a first plurality of indexed layers, and each layer in the first plurality of indexed layers is configured to receive a respective layer input and process the layer input to generate a respective layer output; and one or more subsequent DNNs corresponding to one or more respective machine learning tasks, wherein each subsequent DNN comprises a respective plurality of indexed layers, and each layer in a respective plurality of indexed layers with index greater than one receives input from a preceding layer of the respective subsequent DNN, and one or more preceding layers of respective preceding DNNs, wherein a preceding layer is a layer whose index is one less than the current index.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a whitened neural network layer. One of the methods includes receiving an input activation generated by a layer before the whitened neural network layer in the sequence; processing the received activation in accordance with a set of whitening parameters to generate a whitened activation; processing the whitened activation in accordance with a set of layer parameters to generate an output activation; and providing the output activation as input to a neural network layer after the whitened neural network layer in the sequence.

Abstract: A method of training a machine learning model having multiple parameters, in which the machine learning model has been trained on a first machine learning task to determine first values of the parameters of the machine learning model.

Abstract: A computing device may schedule transmission of software packages on a broadcast/multicast downlink channel. The schedule may also include media transmissions on the channel, and the software package transmissions may be scheduled for times when the media transmissions are using less than or equal to a threshold capacity level of the channel. A software update request may be received from a wireless computing device. Possibly in response to receiving the software update request, a particular software package related to the wireless computing device may be determined. The particular software package may be scheduled to begin transmission on the channel at a particular time. At least an identifier of the channel and the particular time may be transmitted to the wireless computing device.

Abstract: A computing device may schedule transmission of software packages on a broadcast/multicast downlink channel. The schedule may also include media transmissions on the channel, and the software package transmissions may be scheduled for times when the media transmissions are using less than or equal to a threshold capacity level of the channel. A software update request may be received from a wireless computing device. Possibly in response to receiving the software update request, a particular software package related to the wireless computing device may be determined. The particular software package may be scheduled to begin transmission on the channel at a particular time. At least an identifier of the channel and the particular time may be transmitted to the wireless computing device.

Abstract: Methods and systems for performing a sequence of machine learning tasks. One system includes a sequence of deep neural networks (DNNs), including: a first DNN corresponding to a first machine learning task, wherein the first DNN comprises a first plurality of indexed layers, and each layer in the first plurality of indexed layers is configured to receive a respective layer input and process the layer input to generate a respective layer output; and one or more subsequent DNNs corresponding to one or more respective machine learning tasks, wherein each subsequent DNN comprises a respective plurality of indexed layers, and each layer in a respective plurality of indexed layers with index greater than one receives input from a preceding layer of the respective subsequent DNN, and one or more preceding layers of respective preceding DNNs, wherein a preceding layer is a layer whose index is one less than the current index.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing inputs using a neural network system that includes a whitened neural network layer. One of the methods includes receiving an input activation generated by a layer before the whitened neural network layer in the sequence; processing the received activation in accordance with a set of whitening parameters to generate a whitened activation; processing the whitened activation in accordance with a set of layer parameters to generate an output activation; and providing the output activation as input to a neural network layer after the whitened neural network layer in the sequence.

Abstract: A computing device may schedule transmission of software packages on a broadcast/multicast downlink channel. The schedule may also include media transmissions on the channel, and the software package transmissions may be scheduled for times when the media transmissions are using less than or equal to a threshold capacity level of the channel. A software update request may be received from a wireless computing device. Possibly in response to receiving the software update request, a particular software package related to the wireless computing device may be determined. The particular software package may be scheduled to begin transmission on the channel at a particular time. At least an identifier of the channel and the particular time may be transmitted to the wireless computing device.