Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a distilled machine learning model. One of the methods includes training a cumbersome machine learning model, wherein the cumbersome machine learning model is configured to receive an input and generate a respective score for each of a plurality of classes; and training a distilled machine learning model on a plurality of training inputs, wherein the distilled machine learning model is also configured to receive inputs and generate scores for the plurality of classes, comprising: processing each training input using the cumbersome machine learning model to generate a cumbersome target soft output for the training input; and training the distilled machine learning model to, for each of the training inputs, generate a soft output that matches the cumbersome target soft output for the training input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a sequence-to-sequence model that is recurrent in depth while employing self-attention to combine information from different parts of sequences.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output sequence of audio data that comprises a respective audio sample at each of a plurality of time steps. One of the methods includes, for each of the time steps: providing a current sequence of audio data as input to a convolutional subnetwork, wherein the current sequence comprises the respective audio sample at each time step that precedes the time step in the output sequence, and wherein the convolutional subnetwork is configured to process the current sequence of audio data to generate an alternative representation for the time step; and providing the alternative representation for the time step as input to an output layer, wherein the output layer is configured to: process the alternative representation to generate an output that defines a score distribution over a plurality of possible audio samples for the time step.

Abstract: A system comprising a causal convolutional neural network to autoregressively generate a succession of values of a data item conditioned upon previously generated values of the data item. The system includes support memory for a set of support data patches each of which comprises an encoding of an example data item. A soft attention mechanism attends to one or more patches when generating the current item value. The soft attention mechanism determines a set of scores for the support data patches, for example in the form of a soft attention query vector dependent upon the previously generated values of the data item. The soft attention query vector is used to query the memory. When generating the value of the data item at a current iteration layers of the causal convolutional neural network are conditioned upon the support data patches weighted by the scores.

Abstract: This document generally describes a neural network training system, including one or more computers, that trains a recurrent neural network (RNN) to receive an input, e.g., an input sequence, and to generate a sequence of outputs from the input sequence. In some implementations, training can include, for each position after an initial position in a training target sequence, selecting a preceding output of the RNN to provide as input to the RNN at the position, including determining whether to select as the preceding output (i) a true output in a preceding position in the output order or (ii) a value derived from an output of the RNN for the preceding position in an output order generated in accordance with current values of the parameters of the recurrent neural network.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating discrete latent representations of input data items. One of the methods includes receiving an input data item; providing the input data item as input to an encoder neural network to obtain an encoder output for the input data item; and generating a discrete latent representation of the input data item from the encoder output, comprising: for each of the latent variables, determining, from a set of latent embedding vectors in the memory, a latent embedding vector that is nearest to the encoded vector for the latent variable.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing sequences using convolutional neural networks. One of the methods includes, for each of the time steps: providing a current sequence of audio data as input to a convolutional subnetwork, wherein the current sequence comprises the respective audio sample at each time step that precedes the time step in the output sequence, and wherein the convolutional subnetwork is configured to process the current sequence of audio data to generate an alternative representation for the time step; and providing the alternative representation for the time step as input to an output layer, wherein the output layer is configured to: process the alternative representation to generate an output that defines a score distribution over a plurality of possible audio samples for the time step.

Abstract: A computer-implemented method for generating an output token sequence from an input token sequence. The method combines a look ahead tree search, such as a Monte Carlo tree search, with a sequence-to-sequence neural network system. The sequence-to-sequence neural network system has a policy output defining a next token probability distribution, and may include a value neural network providing a value output to evaluate a sequence. An initial partial output sequence is extended using the look ahead tree search guided by the policy output and, in implementations, the value output, of the sequence-to-sequence neural network system until a complete output sequence is obtained.

Abstract: A feedforward generative neural network that generates an output example that includes multiple output samples of a particular type in a single neural network inference. Optionally, the generation may be conditioned on a context input. For example, the feedforward generative neural network may generate a speech waveform that is a verbalization of an input text segment conditioned on linguistic features of the text segment.

Abstract: A computer-implemented method for automatically determining a neural network architecture represents a neural network architecture as a data structure defining a hierarchical set of directed acyclic graphs in multiple levels. Each graph has an input, an output, and a plurality of nodes between the input and the output. At each level, a corresponding set of the nodes are connected pairwise by directed edges which indicate operations performed on outputs of one node to generate an input to another node. Each level is associated with a corresponding set of operations. At a lowest level, the operations associated with each edge are selected from a set of primitive operations. The method includes repeatedly generating new sample neural network architectures, and evaluating their fitness. The modification is performed by selecting a level, selecting two nodes at that level, and modifying, removing or adding an edge between those nodes according to operations associated with lower levels of the hierarchy.

Abstract: There is described a neural network system for generating a graph, the graph comprising a set of nodes and edges. The system comprises one or more neural networks configured to represent a probability distribution over sequences of node generating decisions and/or edge generating decisions, and one or more computers configured to sample the probability distribution represented by the one or more neural networks to generate a graph.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a distilled machine learning model. One of the methods includes training a cumbersome machine learning model, wherein the cumbersome machine learning model is configured to receive an input and generate a respective score for each of a plurality of classes; and training a distilled machine learning model on a plurality of training inputs, wherein the distilled machine learning model is also configured to receive inputs and generate scores for the plurality of classes, comprising: processing each training input using the cumbersome machine learning model to generate a cumbersome target soft output for the training input; and training the distilled machine learning model to, for each of the training inputs, generate a soft output that matches the cumbersome target soft output for the training input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating an output sequence of audio data that comprises a respective audio sample at each of a plurality of time steps. One of the methods includes, for each of the time steps: providing a current sequence of audio data as input to a convolutional subnetwork, wherein the current sequence comprises the respective audio sample at each time step that precedes the time step in the output sequence, and wherein the convolutional subnetwork is configured to process the current sequence of audio data to generate an alternative representation for the time step; and providing the alternative representation for the time step as input to an output layer, wherein the output layer is configured to: process the alternative representation to generate an output that defines a score distribution over a plurality of possible audio samples for the time step.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a sequence to sequence model that is recurrent in depth while employing self-attention to combine information from different parts of sequences.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a neural network. In one aspect, a method includes maintaining data specifying, for each of the network parameters, current values of a respective set of distribution parameters that define a posterior distribution over possible values for the network parameter. A respective current training value for each of the network parameters is determined from a respective temporary gradient value for the network parameter. The current values of the respective sets of distribution parameters for the network parameters are updated in accordance with the respective current training values for the network parameters. The trained values of the network parameters are determined based on the updated current values of the respective sets of distribution parameters.

Abstract: In one aspect, this specification describes a recurrent neural network system implemented by one or more computers that is configured to process input sets to generate neural network outputs for each input set. The input set can be a collection of multiple inputs for which the recurrent neural network should generate the same neural network output regardless of the order in which the inputs are arranged in the collection. The recurrent neural network system can include a read neural network, a process neural network, and a write neural network. In another aspect, this specification describes a system implemented as computer programs on one or more computers in one or more locations that is configured to train a recurrent neural network that receives a neural network input and sequentially emits outputs to generate an output sequence for the neural network input.

Abstract: Methods, systems, and apparatus for classifying a new example using a comparison set of comparison examples. One method includes maintaining a comparison set, the comparison set including comparison examples and a respective label vector for each of the comparison examples, each label vector including a respective score for each label in a predetermined set of labels; receiving a new example; determining a respective attention weight for each comparison example by applying a neural network attention mechanism to the new example and to the comparison examples; and generating a respective label score for each label in the predetermined set of labels from, for each of the comparison examples, the respective attention weight for the comparison example and the respective label vector for the comparison example, in which the respective label score for each of the labels represents a likelihood that the label is a correct label for the new example.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating discrete latent representations of input audio data. Only the discrete latent representation needs to be transmitted from an encoder system to a decoder system in order for the decoder system to be able to effectively to decode, i.e., reconstruct, the input audio data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an encoder neural network that is configured to process an input observation to generate a latent representation of the input observation. In one aspect, a method includes: obtaining a sequence of observations; for each observation in the sequence of observations, processing the observation using the encoder neural network to generate a latent representation of the observation; for each of one or more given observations in the sequence of observations: generating a context latent representation of the given observation; and generating, from the context latent representation of the given observation, a respective estimate of the latent representations of one or more particular observations that are after the given observation in the sequence of observations.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a policy neural network having a plurality of policy parameters and used to select actions to be performed by an agent to control the agent to perform a particular task while interacting with one or more other agents in an environment. In one aspect, the method includes: maintaining data specifying a pool of candidate action selection policies; maintaining data specifying respective matchmaking policy; and training the policy neural network using a reinforcement learning technique to update the policy parameters. The policy parameters define policies to be used in controlling the agent to perform the particular task.