US 12,748,974 B2Granted

Systems and methods for hybrid optimization training of multinomial logit models

AssigneeWalmart Apollo, LLC

InventorsReza Yousefi Maragheh, Ramin Giahi, Aysenur Inan, Hyun Duk Cho, Kaushiki Nag, Sushant Kumar, Kannan Achan

Life of this patent

Estimating the term…

Loading patent…

Abstract

Systems and methods for hybrid optimization of training ranking models is disclosed. A training dataset including a plurality of anchor items, a plurality of recommended item sets, and ground truth data is obtained from a database. A base machine learning model including a step function configured to determine a relevance score is iteratively trained to generate a trained ranking model. The plurality of anchor items and the plurality of recommended item sets are provided as an input to the base machine learning model and the ground truth is provided as a target output. The step function is trained using an adaptive step size according to a first order Barzilai-Borwein (BB) method and a line search method. The trained ranking model is stored in non-transitory memory.

Claims20

Dependent claims

A system, comprising:

a non-transitory memory configured to store a training dataset comprising a plurality of anchor items, a plurality of recommended item sets, and ground truth data;
a processor communicatively coupled to the non-transitory memory, wherein the processor is configured to read a set of instructions to:
obtain, from the non-transitory memory, the training dataset;
obtain a base machine learning model that includes a step function that determines a relevance score;
iteratively train the base machine learning model to generate a trained ranking model, wherein the plurality of anchor items and the plurality of recommended item sets are provided as an input to the base machine learning model and the ground truth data is provided as a target output, and wherein the step function of the base machine learning model is trained using an adaptive step size according to a first order Barzilai-Borwein (BB) process and a line search process;
receive a request for recommended items for a user, wherein the request comprises interaction data characterizing user interactions for the user;
generate, using the trained ranking model, ranking data identifying an ordered set of items based on the interaction data;
transmit the ranking data to a user device for display on a user interface;
receive user interaction data characterizing user interactions between a plurality of users and the user interface;
re-train the trained ranking model using the user interaction data to update the trained ranking model, wherein the step function of the trained ranking model is re-trained using the adaptive step size and the line search process;
and
store the re-trained ranking model in the non-transitory memory.

9 more elements

A computer-implemented method, comprising:

obtaining, from a first database, a training dataset comprising a plurality of anchor items, a plurality of recommended item sets, and ground truth data;
obtaining a base machine learning model that includes a step function that determines a relevance score;
training the base machine learning model to generate a trained ranking model, wherein the plurality of anchor items and the plurality of recommended item sets are provided as an input to the base machine learning model and the ground truth data is provided as a target output, and wherein the step function of the base machine learning model is trained using an adaptive step size;
receiving a request for recommended items for a user, wherein the request comprises interaction data characterizing user interactions for the user;
generating, using the trained ranking model, ranking data identifying an ordered set of items based on the interaction data;
transmit the ranking data to a user device for display on a user interface;
receiving user interaction data characterizing user interactions between a plurality of users and the user interface;
re-training the trained ranking model using the user interaction data to update the trained ranking model, wherein the step function of the trained ranking model is re-trained using the adaptive step size and the line search process;
and
storing the trained ranking model in a second database.

7 more elements

A non-transitory computer-readable medium having instructions stored thereon, wherein the instructions, when executed by at least one processor, cause a device to perform operations comprising:

obtaining, from a first database, a training dataset comprising a plurality of anchor items, a plurality of recommended item sets, and ground truth data;
obtaining a base machine learning model that includes a step function that determines a relevance score;
training the base machine learning model to generate a trained ranking model, wherein the plurality of anchor items and the plurality of recommended item sets are provided as an input to the base machine learning model and the ground truth data is provided as a target output, and wherein the step function of the base machine learning model is trained using an adaptive step size according to a first order Barzilai-Borwein (BB) process and a line search function;
receiving a request for recommended items for a user, wherein the request comprises interaction data characterizing user interactions for the user;
generating, using the trained ranking model, ranking data identifying an ordered set of items based on the interaction data;
transmit the ranking data to a user device for display on a user interface;
receiving user interaction data characterizing user interactions between a plurality of users and the user interface;
re-training the trained ranking model using the user interaction data to update the trained ranking model, wherein the step function of the trained ranking model is re-trained using the adaptive step size and the line search process;
and
storing the trained ranking model in a second database.

7 more elements

Dependencies are inferred from the claim text.