Alibaba Group Holding Limited

Advanced New Technologies Co., Ltd.

Alipay (Hangzhou) Information Technology Co., Ltd.

Alipay Labs (Singapore) Pte. Ltd.

Advantageous New Technologies Co., Ltd.

Feng Qi

Zhigang Hua

Xingwen Zhang

Shuanghong Yang

Shuang Yang

Jia Yan

Yingping Cao

Dapeng Fu

Gan Liu

Dingxiang Hu

Yingqi Liu

Disclosed methods, systems, and apparatus, include computer programs encoded on computer storage media, for performing allocation of M resources among N users into K pools by solving a knapsack problem (KP) using a distributed computing system. The method includes: receiving data representing K global constraints and L local constraints of the KP; transforming the KP into a dual problem using K dual multipliers; decomposing the dual problem into N sub-problems; performing two or more iterations in solving the dual problem, wherein in one iteration, for each dual multiplier corresponding to a global constraint corresponding to a pool: determining an updated dual multiplier for the global constraint corresponding to the pool to be a non-negative threshold; and computing M decision variables of each of the N users corresponding to the updated dual multiplier in solving each of the N sub-problems corresponding to the each of the N users.

Distributed resource allocation

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for grouping users are provided. One embodiment of the methods includes: dividing a plurality of users targeted by the an advertisement candidate into a plurality of user buckets, wherein each of the plurality of user buckets is associated with a first conversion score; obtaining a trained prediction model corresponding to the advertisement, wherein the trained prediction model is able to predict a conversion score based at least on the first conversion score associated with a user bucket and a second conversion score associated with a group of user buckets comprising the user bucket; and constructing an optimization model using the trained prediction model, wherein an objective function of the optimization problem is to maximize a total conversion score with a grouping strategy determined by solving the optimization problem.

Method and system for optimizing user grouping for advertisement

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for personalizing offers are provided. One of the methods includes: collecting response data comprising one or more offers made to each of a plurality of users of a platform and one or more corresponding responses, wherein the one or more offers are from a group of offer choices; creating a training dataset comprising the collected response data and one or more features associated with each of the plurality of users; training a machine learning model using the training dataset, wherein the trained machine learning model is configured to predict the plurality of users' responses to future offers; obtaining a plurality of projected profits for the platform using the trained machine learning model, wherein each of the plurality of projected profits corresponds to making one of the group of the predetermined offers to one of the plurality of users.

Method and system for personalizing offers

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for evaluating risk are provided. One of the methods includes: obtaining a plurality of transactions each comprising a plurality of data dimensions, wherein some of the plurality of transactions are labeled as risky transactions and some of the plurality of transactions are labeled as safe transactions; obtaining at least one of the plurality of data dimensions as an output space and the plurality of data dimensions other than the at least one data dimension as an input space; initializing a first mapping from the input space to a latent space and a second mapping from the latent space to the output space, wherein the first mapping comprises mapping the input space to the latent space according to an inverse of a generalized linear model; and optimizing the first mapping and the second mapping to generate a Bezier surface.

System and method for evaluating risk

Disclosed herein are methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing allocation of M resources among N users into K pools by solving a knapsack problem (KP) using a distributed computing system that includes a number of individual solvers. The method includes: receiving data representing K global constraints and L local constraints of the KP; decomposing the KP into N sub-problems using K dual multipliers, each of the N sub-problems corresponding to a respective one of the N users and subject to the L local constraints w.r.t. the corresponding user, wherein N is in an order of billions or larger; determining the number of individual solvers for solving the N sub-problems; distributing the N sub-problems among the number of individual solvers; and solving the KP by the distributed computing system by performing two or more iterations.

Disclosed herein are methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing allocating M resources subject to L constraints. The method includes: receiving data representing L constraints, wherein each of the L constraints corresponds to a subset of M resources and restricts a respective maximum number C of resources to be selected among the subset of the M resources, wherein the L constraints has a hierarchal structure; determining a topological ordering of the L constraints; selecting all the M resources as an initial selection; removing resources from the initial selection by traversing each constraint in the topological ordering of the L constraints; and allocating the selected resources after traversing all the L constraints in the topological ordering of the L constraints.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for optimizing resource allocation are provided. One of the methods includes: processing a plurality of first objective functions in parallel to determine a plurality of allocation plans, wherein each of the allocation plans corresponds to allocating zero or more of a plurality of resources associated with a platform to a user, and the plurality of first objective functions share one or more dual multipliers; determining a plurality of profits and costs for the platform in parallel based on the plurality of the allocation plans; aggregating the calculated profits and costs using parallel reduction; updating the one or more dual multipliers based on the aggregated costs to determine whether an exit condition is satisfied; if the exit condition is not satisfied, repeating the processing the plurality of the first objective functions based on the updated dual multipliers.

Growing community of inventors

Hangzhou, China

Feng Qi