Abstract: Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for data communication with differentially private intersection dataset. On example method includes: determining a plurality of bias parameters based on a parameter for establishing a differential privacy (DP) data protection; determining an intersection dataset based on client data from a client device and server data from the server device; updating the intersection dataset based on the client data and the plurality of bias parameters; and returning the updated intersection data having the DP data protection to the client device.

Abstract: Embodiments of this specification provide methods, apparatuses systems, and computer-readable media for data privacy-preserving training of a service prediction model. In an example training process, a member device performs prediction by using the service prediction model and object feature data held by the member device, and determines, by using a prediction result, update parameters used to update model parameters, where the update parameters include sub-parameters for computational layers of the service prediction model; and divides the computational layers into first-type and second-type computational layers using the sub-parameters; and performs privacy processing on sub-parameters of the first-type computational layers, and outputs processed sub-parameters. Processed sub-parameters of member devices can be aggregated into aggregated sub-parameters.

Abstract: Implementations of the present specification provide a query processing method and apparatus. The method includes: first, determining query types of L queries to be performed on a target dataset, where the target dataset includes data of an object; next, determining query sensitivity of each query type of the query types for the target dataset; and then, determining, based on the query sensitivity corresponding to each query and a privacy budget parameter predetermined for a total set of the L queries, a noise power allocated to each query. Based on this, for a target query in the L queries, an actually returned result of the target query can be determined as an original query result of the target query added with a target noise sampled from target noise distribution of differential privacy, where the target noise distribution is determined based on the noise power allocated to the target query. As such, privacy of the target dataset can be protected.

Abstract: Implementations of this specification provide query optimization methods, apparatuses, and systems for secure multi-party databases. In an implementation, a method includes: receiving a current query associated with a plurality of target database of a multi-party database system, generating a plurality of execution plans for the current query, determining, for each execution plan, a respective cost computation formula of a plurality of cost computation values for computing an execution cost of jointly executing the execution plan by the plurality of target databases, receiving a secure computation result from each of a plurality of query engines corresponding to the plurality of target databases, and determining an optimal execution plan having a lowest cost value in the plurality of cost computation formulas based on the secure computation result.

Abstract: This specification provides example data query methods, apparatuses, and systems for a multi-party secure database. In an example computer-implemented method, a central node receives a data query request from a data requester, where the data query request relates to one or more pieces of target data stored at one or more data providers. The central node determines whether the data query request satisfies respective security requirements of the one or more pieces of target data. In response to determining that the data query request satisfies respective security requirements of the one or more pieces of target data, the central node processes the one or more pieces of target data to obtain result data; and sends the result data to the data requester. The central node can restrict query of the target data by the data requester based on the security requirement of the target data stored at the data provider.

Abstract: This specification provides example computer-implemented methods and apparatuses for jointly updating a service model based on privacy protection. In an example iteration process, a serving party provides, to each data party, global model parameters and a mapping relationship between the data party and N parameter groups obtained by dividing the global model parameters. Each data party updates a local service model by using the global model parameters, and further updates an updated local service model based on local service data, to upload model parameters in a new service model in a parameter group corresponding to the data party to the serving party. Then, the serving party successively fuses received parameter groups to update the global model parameters.

Abstract: Embodiments of this specification provide computer-implemented methods, apparatuses, and systems for training a model using multiple data owners. In an example method, a second data owner determines, according to first data, second feature data intersecting the first data. The following main iteration process is performed until an iteration end condition is met: performing, for each training unit by using a first training sample and a second training sample, cooperative training on a first model, a second model, and a third model that participate in training of the training unit. A master server performs federated aggregation on the trained first model and/or third model in each training unit, to obtain a corresponding first global model and/or third global model. The first model is updated according to the first global model and/or the third model is updated according to the third global model.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for federal learning are disclosed. One exemplary method may include receiving, by a client device from a server device, model parameters of a global machine learning model being collaboratively trained by the client device and the server device; constructing, by the client device, a local machine learning model based on the model parameters of the global machine learning model, wherein the local machine learning model comprises two branches corresponding to two loss functions; training, by the client device, the local machine learning model based on local training data, wherein the training comprises updating the model parameters to minimize a first loss function of the two loss functions and maximize a second loss function of the two loss functions; and sending, by the client device, the updated model parameters back to the server device.

Abstract: Information regarding display of selected content items in assigned slots is provided. The information is based on a selected template that includes areas for groups of slots for related content items, such as content items that are targeted to a common targeting object. Candidate content items, identified based on targeting or other distribution information, are assigned to selected slots to determine at least part of the information regarding display of the selected content items.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for federal learning with differentially private (DP) intrinsic quantization are disclosed. One exemplary method may include obtaining a parameter vector of a local model; updating the parameter vector of the local model by adding a plurality of noise vectors to the parameter vector of the local model; performing quantization to the updated parameter vector to obtain a quantized parameter vector, wherein the quantization maps coordinates in the updated parameter vector to a set of discrete finite values; and transmitting, to a server, the quantized parameter vector and at least one of the plurality of noise vectors for the server to update a global model.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on computer storage media, for federal learning with differentially private (DP) intrinsic quantization are disclosed. One exemplary method may include obtaining a parameter vector of a local model; updating the parameter vector of the local model by adding a plurality of noise vectors to the parameter vector of the local model; performing quantization to the updated parameter vector to obtain a quantized parameter vector, wherein the quantization maps coordinates in the updated parameter vector to a set of discrete finite values; and transmitting, to a server, the quantized parameter vector and at least one of the plurality of noise vectors for the server to update a global model.

Abstract: Information regarding display of selected content items in assigned slots is provided. The information is based on a selected template that includes areas for groups of slots for related content items, such as content items that are targeted to a common targeting object. Candidate content items, identified based on targeting or other distribution information, are assigned to selected slots to determine at least part of the information regarding display of the selected content items.

Abstract: A search system server routes a given query to devices that create a response to the query. For query processing, a mixer server determines a sequence in which various servers are called, formats requests, and aggregates data obtained from the various servers. A query reviser creates various different queries that differ from, but are based on, the given query and prior queries submitted by a same user that submitted the given query or location information associated with the user. Portions of content are incorporated into the search results page including content selected based on the given query and based on the various different queries. A visual representation of the various portions of content within the search results page is generated. A display of a client device that submitted the given query is updated to present the search results page including the visual representation of the various portions of content.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for enhancing selecting relevant and diverse advertisements. In one aspect, a method includes receiving an initial query, selecting one or more additional queries relating to the initial query, including selecting additional queries having a greatest commerciality, identifying one or more content items for each of the additional queries, the one or more content items forming a content block, and providing a content block and an associated additional query to a client device to be displayed along with search results associated with the initial query.

Abstract: The present disclosure discloses a system and method for deploying customized machine learning services. Specifically, a computing device of a machine learning service provider maintains a meta training component and a plurality of meta data schemas. The computing device also generates a customized data schema based on customization of one of the plurality of meta data schemas by a machine learning service client. Further, the computing device can generate a training component based on the meta training component. The training component is compatible with the customized data schema. Then, the computing device deploys the customized data schema and the training component to one or more client devices to automatically generate a machine learning model.

Abstract: A search system server routes a given query to devices that create a response to the query. For query processing, a mixer server determines a sequence in which various servers are called, formats requests, and aggregates data obtained from the various servers. A query reviser creates various different queries that differ from, but are based on, the given query and prior queries submitted by a same user that submitted the given query or location information associated with the user. Portions of content are incorporated into the search results page including content selected based on the given query and based on the various different queries. A visual representation of the various portions of content within the search results page is generated. A display of a client device that submitted the given query is updated to present the search results page including the visual representation of the various portions of content.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for enhancing selecting relevant and diverse advertisements. In one aspect, a method includes receiving an initial query, selecting one or more additional queries relating to the initial query, including selecting additional queries having a greatest commerciality, identifying one or more content items for each of the additional queries, the one or more content items forming a content block, and providing a content block and an associated additional query to a client device to be displayed along with search results associated with the initial query.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for enhancing selecting relevant and diverse advertisements. In one aspect, a method includes receiving an initial query, selecting one or more additional queries relating to the initial query, including selecting a first additional query and a second diverse additional query, identifying one or more content items for each of the additional queries, the one or more content items forming a content block, and providing a content block and associated additional query to be displayed at a client device along with search results associated with the initial query.

Abstract: A method and system for identifying information about people is provided. The information system identifies groups of people that have relationships based on their relationships to documents or more generally to objects. The information system initially is provided with an indication of which people have which relationships to which documents. The information system then identifies clusters of people based on having a relationship to the same objects. The information system may also identify clusters of related objects associated with a cluster of people. When a user wants to identify information about a person, the user can provide the name of that person to the information system. The information system then can retrieve and display the names of the other people who are in the same cluster as the person.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for enhancing selecting relevant and diverse advertisements. In one aspect, a method includes receiving an initial query, selecting one or more additional queries relating to the initial query, identifying one or more content items for each query, forming a content block, balancing each content block, and providing, for each query, a content block along with search results associated with the initial query for display or other processing by a device.