Abstract: Implementations of this specification provide prediction methods and apparatuses for adjusting computing power. One method comprises receiving a prediction request, wherein the prediction request comprises a sample to be tested, determining a computing power coefficient allocated to the prediction request, wherein the computing power coefficient indicates a proportion of hardware computing power resources allocated to the prediction request to total hardware computing power resources needed for a neural network model to run on a computing platform, determining k sub-networks inn sub-networks of the neural network model to be used for a present time based on the computing power coefficient, where n>2, and inputting the sample to be tested to the k sub-networks to obtain a prediction result.

Abstract: A multi-booth traffic allocation scheme for multiple users if provided. In an implementation, an exposure probability of each booth and a unit traffic revenue and corresponding unit resource consumption of each candidate display object for each user are obtained. An optimization objective and a resource consumption constraint corresponding to a multi-booth traffic allocation scheme are constructed. The constructed optimization objective includes a sum value of traffic revenues for each user under the multi-booth traffic allocation scheme. The traffic revenues for each user are determined based on the exposure probability of each booth, a booth allocation scheme of candidate display objects, and the unit traffic revenue of each candidate display object. A multi-booth traffic allocation scheme when the resource consumption constraint is satisfied and the optimization objective is maximized is determined.

Abstract: A target sample including a sample feature, a first label, and a second label is obtained, where a user corresponds to the target sample, the first label indicates whether a target object is clicked on by the user, and the second label indicates whether the user implements a target behavior related to the target object. Model processing is performed on the sample feature using a prediction model with a first branch and first probability and a second branch and second probability. A first loss is determined based on a first label value and the first probability. When a predetermined condition is satisfied, the second loss is determined based on a second label value of the second label and the second probability, and a predicted loss of the target sample is determined based on the first loss and the second loss. Training the prediction model based on the predicted loss.