Abstract: This application describes a model training method, applied to the field of artificial intelligence. The method includes a computing core of a first processor obtains an embedding used for model training, and writes an updated embedding to a first memory of the first processor instead of transferring the updated embedding to a second processor after model training is completed. In this application, after updating an embedding, the first processor saves the updated embedding to the first memory of the first processor. Without needing to wait for the second processor to complete a process of transferring a second target embedding to a GPU, the first processor may directly obtain the updated embedding and perform model training of a next round based on the updated embedding, provided that the first processor may obtain a latest updated embedding.

Abstract: The method includes: collecting historical operations of sample users for M items, and predicting a preference value of a target user for each of the M items according to historical operations of the sample users for each of the M items, collecting classification data of N to-be-recommended items, and classifying the N to-be-recommended items according to the classification data of the N to-be-recommended items, to obtain X themes, where each of the X themes includes at least one of the N to-be-recommended items, and the N to-be-recommended items are some or all of the M items; calculating a preference value of the target user for each of the X themes according to a preference value of the target user for a to-be-recommended item included in each of the X themes; and pushing a target theme to the target user.

Abstract: Embodiments of this application provide a data processing method and apparatus to better learn a vector representation value of each feature value in a continuous feature. The method specifically includes: The data processing apparatus obtains the continuous feature from sample data, and then performs discretization processing on the continuous feature by using a discretization model, to obtain N discretization probabilities corresponding to the continuous feature. The N discretization probabilities correspond to N preset meta-embeddings, and N is an integer greater than 1. Finally, the data processing apparatus determines a vector representation value of the continuous feature based on the N discretization probabilities and the N meta-embeddings.

Abstract: The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m?3, and m>n?2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data.

Abstract: The method includes: collecting historical operations of sample users for M items, and predicting a preference value of a target user for each of the M items according to historical operations of the sample users for each of the M items, collecting classification data of N to-be-recommended items, and classifying the N to-be-recommended items according to the classification data of the N to-be-recommended items, to obtain X themes, where each of the X themes includes at least one of the N to-be-recommended items, and the N to-be-recommended items are some or all of the M items; calculating a preference value of the target user for each of the X themes according to a preference value of the target user for a to-be-recommended item included in each of the X themes; and pushing a target theme to the target user.

Abstract: System and method for training a recommender system (RS). The RS is configured to make recommendations in respect of a bipartite graph that comprises a plurality of user nodes, a plurality of item nodes, and an observed graph topology that defines edges connecting at least some of the user nodes to some of the item nodes, the RS including an existing graph neural network (GNN) model configured by an existing set of parameters. The method includes: applying a loss function to compute an updated set of parameters for an updated GNN model that is trained with a new graph using the first set of parameters as initialization parameters, the loss function being configured to distil knowledge based on node embeddings generated by the existing GNN model in respect of an existing graph, wherein the new graph includes a plurality of user nodes and a plurality of item nodes that are also included in the existing graph; and replacing the existing GNN model of the RS with the updated GNN model.

Abstract: Method and system for processing a bipartite graph that comprises a plurality of first nodes of a first node type, and a plurality of second nodes of a second type, comprising: generating a target first node embedding for a target first node based on features of second nodes and first nodes that are within a multi-hop first node neighbourhood of the target first node, the target first node being selected from the plurality of first nodes of the first node type; generating a target second node embedding for a target second node based on features of first nodes and second nodes that are within a multi-hop second node neighbourhood of the target second node, the target second node being selected from the plurality of second nodes of the second node type; and determining a relationship between the target first node and the target second node based on the target first node embedding and the target second node embedding.

Abstract: A recommendation method includes generating a feature sequence based on to-be-predicted data of a user for a target object and according to a preset encoding rule, obtaining probability distribution information corresponding to each feature in the feature sequence, and obtaining, through calculation, a feature vector corresponding to each feature, obtaining a predicted score of the user for the target object based on values of N features and a feature vector corresponding to each of the N features, and recommending the target object to the user when the predicted score is greater than or equal to a preset threshold.

Abstract: This application relates to the field of artificial intelligence.

Abstract: A data processing method related to the field of artificial intelligence includes adding an architecture parameter to each feature interaction item in a first model, to obtain a second model, where the first model is a factorization machine (FM)-based model, and the architecture parameter represents importance of a corresponding feature interaction item; performing optimization on architecture parameters in the second model to obtain the optimized architecture parameters; and obtaining, based on the optimized architecture parameters and the first model or the second model, a third model through feature interaction item deletion.

Abstract: Example user behavior prediction methods and apparatus are described. One example method includes obtaining a first contribution value of each piece of characteristic data for a specified behavior after obtaining behavior prediction information including a plurality of pieces of characteristic data. Every N pieces of characteristic data in the plurality of pieces of characteristic data may be processed by using one corresponding characteristic interaction model, to obtain a second contribution value of the every N pieces of characteristic data for the specified behavior. Finally, an execution probability of executing the specified behavior by a user may be determined based on the obtained first contribution value and the obtained second contribution value, to predict a user behavior. In the example method, interaction impact of the plurality of pieces of characteristic data on the specified behavior is considered during behavior prediction.

Abstract: The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m?3, and m>n?2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data.

Abstract: A recommendation model training method, a selection probability prediction method, and an apparatus are provided. The training method includes obtaining a training sample, where the training sample includes a sample user behavior log, position information of a sample recommended object, and a sample label. The training method further includes performing joint training on a position aware model and a recommendation model by the training sample, to obtain a trained recommendation model, where the position aware model predicts probabilities that a user pays attention to a target recommended object when the target recommended object is at different positions, and the recommendation model predicts, when the user pays attention to the target recommended object, a probability that the user selects the target recommended object.

Abstract: The method includes: obtaining a plurality of pieces of feature data; automatically performing two different types of nonlinear combination processing operations on the plurality of pieces of feature data to obtain two groups of processed data, where the two groups of processed data include a group of higher-order data and a group of lower-order data, the higher-order data is related to a nonlinear combination of m pieces of feature data in the plurality of pieces of feature data, and the lower-order data is related to a nonlinear combination of n pieces of feature data in the plurality of pieces of feature data, where m?3, and m>n?2; and determining prediction data based on a plurality of pieces of target data, where the plurality of pieces of target data include the two groups of processed data.

Abstract: An application sorting method and apparatus are provided. The method includes: obtaining, a positive operation probability and positive operation feedback information of each of at least two data samples; calculating an uncertainty parameter of a positive operation probability of a first data sample based on the positive operation probabilities and the positive operation feedback information of the at least two data samples and feature indication information of at least one same feature in a plurality of features in the at least two data samples; and correcting the positive operation probability of the first data sample by using the uncertainty parameter of the positive operation probability; and sorting, based on corrected positive operation probabilities, application programs corresponding to the at least two data samples.

Abstract: A recommendation model training method includes selecting a positive sample in a sample set, and adding the positive sample to a training set, where the sample set includes the positive sample and negative samples, each sample includes n sample features, n?1, and the sample features of each sample include a feature used to represent whether the sample is a positive sample or a negative sample, calculating sampling probabilities of the negative samples in the sample set by using a preset algorithm, selecting a negative sample from the sample set based on the sampling probability, and adding the negative sample to the training set, and performing training by using the samples in the training set, to obtain a recommendation model.

Abstract: In a recommendation-providing method in the field of artificial intelligence, an apparatus for generating recommendations obtains a recommendation system status parameter based on a plurality of historical recommended objects and a user behavior for each historical recommended object, such as clicks or downloads. The apparatus determines a target set among lower-level sets according to the recommendation system status parameter and a selection policy corresponding to an upper-level set, where the lower-level sets and upper-level set correspond to nodes on a clustering tree representing available to-be-presented objects, and each set corresponds to one selection policy. The apparatus then determines a target to-be-recommended object from the to-be recommended objects in the target set.

Abstract: A recommendation method includes generating a feature sequence based on to-be-predicted data of a user for a target object and according to a preset encoding rule, obtaining probability distribution information corresponding to each feature in the feature sequence, and obtaining, through calculation, a feature vector corresponding to each feature, obtaining a predicted score of the user for the target object based on values of N features and a feature vector corresponding to each of the N features, and recommending the target object to the user when the predicted score is greater than or equal to a preset threshold.

Abstract: Example user behavior prediction methods and apparatus are described. One example method includes obtaining a first contribution value of each piece of characteristic data for a specified behavior after obtaining behavior prediction information including a plurality of pieces of characteristic data. Every N pieces of characteristic data in the plurality of pieces of characteristic data may be processed by using one corresponding characteristic interaction model, to obtain a second contribution value of the every N pieces of characteristic data for the specified behavior. Finally, an execution probability of executing the specified behavior by a user may be determined based on the obtained first contribution value and the obtained second contribution value, to predict a user behavior. In the example method, interaction impact of the plurality of pieces of characteristic data on the specified behavior is considered during behavior prediction.

Abstract: A route planning method includes obtaining exercise capability information of a wearer and one or more candidate routes, where the candidate routes include attribute features that comprise historical exercise capability information, where the historical exercise capability information is information calculated according to a first preset rule and based on obtained exercise capability information of a plurality of users having exercised along the candidate routes; determining a target route based on the attribute features of the candidate routes and the exercise capability information of the wearer; and outputting the target route information.