Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for training a neural network system to perform multiple machine learning tasks.

Abstract: A method of accessing data includes storing a table that includes a plurality of tablets corresponding to distinct non-overlapping table portions. Respective pluralities of tablet access objects and application objects are stored in a plurality of servers. A distinct application object and distinct tablet are associated with each tablet access object. Each application object corresponds to a distinct instantiation of an application associated with the table. The tablet access objects and associated application objects are redistributed among the servers in accordance with a first load-balancing criterion. A first request directed to a respective tablet is received from a client. In response, the tablet access object associated with the respective tablet is used to perform a data access operation on the respective tablet, and the application object associated with the respective tablet is used to perform an additional computational operation to produce a result to be returned to the client.

Abstract: A method for determining a placement for machine learning model operations across multiple hardware devices is described.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for predicting likelihoods of conditions being satisfied using recurrent neural networks. One of the systems is configured to process a temporal sequence comprising a respective input at each of a plurality of time steps and comprises: one or more recurrent neural network layers; one or more logistic regression nodes, wherein each of the logistic regression nodes corresponds to a respective condition from a predetermined set of conditions, and wherein each of the logistic regression nodes is configured to, for each of the plurality of time steps: receive the network internal state for the time step; and process the network internal state for the time step in accordance with current values of a set of parameters of the logistic regression node to generate a future condition score for the corresponding condition for the time step.

Abstract: A method for deleting obsolete files from a file system is provided. The method includes receiving a request to delete a reference to a first target file of a plurality of target files stored in a file system, the first target file having a first target file name. A first reference file whose file name includes the first target file name is identified. The first reference file is deleted from the file system. The method further includes determining whether the file system includes at least one reference file, distinct from the first reference file, whose file name includes the first target file name. In accordance with a determination that the file system does not include the at least one reference file, the first target file is deleted from the file system.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for receiving a request from a client to process a computational graph; obtaining data representing the computational graph, the computational graph comprising a plurality of nodes and directed edges, wherein each node represents a respective operation, wherein each directed edge connects a respective first node to a respective second node that represents an operation that receives, as input, an output of an operation represented by the respective first node; identifying a plurality of available devices for performing the requested operation; partitioning the computational graph into a plurality of subgraphs, each subgraph comprising one or more nodes in the computational graph; and assigning, for each subgraph, the operations represented by the one or more nodes in the subgraph to a respective available device in the plurality of available devices for operation.

Abstract: Systems and methods for processing loops in computational graphs representing machine learning models are disclosed. An example method begins with obtaining data representing a computational graph. Data identifying an allocation of the computational graph across devices is obtained. Additionally, one or more nodes in the computational graph that represent a respective control flow statement are identified. For each identified node, a structure of nodes and edges that represents an operation that provides a current state of recursion or iteration in the respective control flow statement is generated. This structure is inserted into the computational graph and the allocation of nodes to devices is modified to assign the structure to a device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a computer chip floorplan. One of the methods includes obtaining netlist data for a computer chip; and generating a computer chip floorplan, comprising placing a respective node at each time step in a sequence comprising a plurality of time steps, the placing comprising, for each time step: generating an input representation for the time step; processing the input representation using a node placement neural network having a plurality of network parameters, wherein the node placement neural network is configured to process the input representation in accordance with current values of the network parameters to generate a score distribution over a plurality of positions on the surface of the computer chip; and assigning the node to be placed at the time step to a position from the plurality of positions using the score distribution.

Abstract: A method performs large-scale data processing in a distributed and parallel processing environment. The method defines application-independent map and reduce operations, each invoking one or more library functions that automatically handle data partitioning, parallelization of computations, and fault tolerance. A user specifies a map operation, which calls one or more of the application-independent map operators to perform data read and write operations. A user also specifies a reduce operation, which calls one or more of the application-independent reduce operators to perform data read and write operations. The method executes application-independent map worker processes. Each map worker process executes the user-specified map operation to read designated portions of input files and store intermediate data values in intermediate data structures. The method also executes application-independent reduce worker processes.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a computer chip placement. One of the methods includes obtaining netlist data for a computer chip; and generating a computer chip placement, comprising placing a respective macro node at each time step in a sequence comprising a plurality of time steps, the placing comprising, for each time step: generating an input representation for the time step; processing the input representation using a node placement neural network having a plurality of network parameters, wherein the node placement neural network is configured to process the input representation in accordance with current values of the network parameters to generate a score distribution over a plurality of positions on the surface of the computer chip; and assigning the macro node to be placed at the time step to a position from the plurality of positions using the score distribution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for distributing machine learning workloads, e.g., computations for training a neural network or computing an inference using a neural network, across multiple hardware accelerators.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating a computer chip placement. One of the methods includes obtaining netlist data for a computer chip; and generating a computer chip placement, comprising placing a respective macro node at each time step in a sequence comprising a plurality of time steps, the placing comprising, for each time step: generating an input representation for the time step; processing the input representation using a node placement neural network having a plurality of network parameters, wherein the node placement neural network is configured to process the input representation in accordance with current values of the network parameters to generate a score distribution over a plurality of positions on the surface of the computer chip; and assigning the macro node to be placed at the time step to a position from the plurality of positions using the score distribution.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for distributing machine learning workloads, e.g., computations for training a neural network or computing an inference using a neural network, across multiple hardware accelerators.

Abstract: A method performs large-scale data processing in a distributed and parallel processing environment. The method defines application-independent map and reduce operations, each invoking one or more library functions that automatically handle data partitioning, parallelization of computations, and fault tolerance. A user specifies a map operation, which calls one or more of the application-independent map operators to perform data read and write operations. A user also specifies a reduce operation, which calls one or more of the application-independent reduce operators to perform data read and write operations. The method executes application-independent map worker processes. Each map worker process executes the user-specified map operation to read designated portions of input files and store intermediate data values in intermediate data structures. The method also executes application-independent reduce worker processes.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for distributing machine learning workloads, e.g., computations for training a neural network or computing an inference using a neural network, across multiple hardware accelerators.

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: A method for determining a placement for machine learning model operations across multiple hardware devices includes receiving data specifying machine learning operations, and determining a placement that assigns each of the operations specified by the data to a respective device from the multiple hardware devices.

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: A method of accessing data includes storing a table that includes a plurality of tablets corresponding to distinct non-overlapping table portions. Respective pluralities of tablet access objects and application objects are stored in a plurality of servers. A distinct application object and distinct tablet are associated with each tablet access object. Each application object corresponds to a distinct instantiation of an application associated with the table. The tablet access objects and associated application objects are redistributed among the servers in accordance with a first load-balancing criterion. A first request directed to a respective tablet is received from a client. In response, the tablet access object associated with the respective tablet is used to perform a data access operation on the respective tablet, and the application object associated with the respective tablet is used to perform an additional computational operation to produce a result to be returned to the client.

Abstract: The usefulness of content (target content), such as advertisements, may be increased by determining additional content and providing such additional content in association with the content. The target content may be text, a Web page, a URL, a search query, etc. The additional content might be related suggested queries (e.g. “Try a search for ______”), news articles (or excerpts or summaries thereof), reviews (or excerpts or summaries thereof), advertisements, user group messages, etc.