公开(公告)号：US20250077959A1

公开(公告)日：2025-03-06

申请号：US18461703

申请日：2023-09-06

Applicant: AMPERITY, INC.

Inventor： Yan YAN ,  Pranav Behari LAL ,  Nicholas RESNICK ,  Joyce GORDON

IPC: G06N20/00

Abstract: In some implementations, the techniques described herein relate to a method including: loading a current and a new model, the new model including the most recent version of the current model; computing a migration duration based on computed properties, namely the jitter in predictions between the current and the new models based on imputing the same inference data to both models; blending outputs of the current model with outputs of the new model according to weights computed for a current time step in the migration process; and serving new predictions using the new model when the migration duration expires.

公开(公告)号：US20250148322A1

公开(公告)日：2025-05-08

申请号：US19017142

申请日：2025-01-10

Applicant: AMPERITY, INC.

Inventor： Yan YAN ,  Aria HAGHIGHI ,  Nicholas RESNICK ,  Andrew LIM

IPC: G06N5/04 ,  G06F16/23 ,  G06F16/24 ,  G06N20/00 ,  G06Q30/01 ,  G06Q30/0201 ,  G06Q30/0202

Abstract: Disclosed are techniques for generating features to train a predictive model to predict a customer lifetime value or churn rate. In one embodiment, a method is disclosed comprising receiving a record that includes a plurality of fields and selecting a value associated with a selected field in the plurality of fields. The method then queries a lookup table comprising a mapping of values to aggregated statistics using the value and receives an aggregated statistic based on the querying. Next, the method generates a feature vector by annotating the record with the aggregated statistic. The method uses this feature vector as an input to a predictive model.

公开(公告)号：US20230131884A1

公开(公告)日：2023-04-27

申请号：US17511946

申请日：2021-10-27

Applicant: AMPERITY, INC.

Inventor： Andrew LIM ,  Joseph CHRISTIANSON ,  Joyce GORDON ,  Nicholas RESNICK ,  Yan YAN

IPC: G06N7/00 ,  G06N20/20

Abstract: The example embodiments are directed toward improvements in generating affinity groups. In an embodiment, a method is disclosed comprising generating probabilities of object interactions for a plurality of users, a given object recommendation ranking for a respective user comprising a ranked list of object attributes; calculating interaction probabilities for each user over a forecasting window; calculating affinity group rankings based on the probabilities of object interactions and the interaction probabilities for each user; and grouping the plurality of users based on the affinity group rankings.

公开(公告)号：US20240152782A1

公开(公告)日：2024-05-09

申请号：US18390803

申请日：2023-12-20

Applicant: AMPERITY, INC.

Inventor： Yan YAN ,  Aria HAGHIGHI ,  Nicholas RESNICK ,  Andrew LIM

IPC: G06N5/04 ,  G06F16/23 ,  G06F16/24 ,  G06N20/00 ,  G06Q30/0201

CPC classification number: G06N5/04 ,  G06F16/2379 ,  G06F16/24 ,  G06N20/00 ,  G06Q30/0201 ,  G06Q30/0202

Abstract: Disclosed are techniques for generating features to train a predictive model to predict a customer lifetime value or churn rate. In one embodiment, a method is disclosed comprising receiving a record that includes a plurality of fields and selecting a value associated with a selected field in the plurality of fields. The method then queries a lookup table comprising a mapping of values to aggregated statistics using the value and receives an aggregated statistic based on the querying. Next, the method generates a feature vector by annotating the record with the aggregated statistic. The method uses this feature vector as an input to a predictive model.

公开(公告)号：US20230252503A1

公开(公告)日：2023-08-10

申请号：US17854154

申请日：2022-06-30

Applicant: AMPERITY, INC.

Inventor： Joyce GORDON ,  Pranav Behari LAL ,  Nicholas RESNICK ,  James WU ,  Yan YAN

IPC: G06Q30/02 ,  G06N5/00

CPC classification number: G06Q30/0202 ,  G06N5/003

Abstract: In some aspects, the techniques described herein relate to a method including: receiving a vector, the vector including a plurality of features related to a user; predicting a return probability for the user based on the vector using a first predictive model; adjusting the return probability using a fitted sigmoid function to generate an adjusted return probability; and predicting a lifetime value of the user using the adjusted return probability and at least one other prediction by combining the adjusted return probability and the at least one other prediction.

公开(公告)号：US20230128579A1

公开(公告)日：2023-04-27

申请号：US17511747

申请日：2021-10-27

Applicant: AMPERITY, INC.

Inventor： Nicholas RESNICK ,  Joseph CHRISTIANSON ,  Joyce GORDON ,  Andrew LIM ,  Yan YAN

IPC: G06N3/08 ,  G06N3/04 ,  G06N5/00

Abstract: The example embodiments are directed toward predicting the lifetime value of a user using an ensemble model. In an embodiment, a system is disclosed, including a generative model for generating a first prediction representing a first lifetime value of a user during a forecasting period and a discriminative model configured for generating a second prediction representing a second lifetime value of the user during the forecasting period. The system further includes a meta-model for receiving the first prediction and the second prediction and generating a third prediction based on the first prediction and the second prediction, the third prediction representing a third lifetime value of the user during the forecasting period.

公开(公告)号：US20250068653A1

公开(公告)日：2025-02-27

申请号：US18805907

申请日：2024-08-15

Applicant: AMPERITY, INC.

Inventor： Yan YAN ,  Nicholas RESNICK ,  Jean RUGGIERO ,  Joseph CHRISTIANSON

IPC: G06F16/28

Abstract: The disclosed embodiments relate to devices, computer-readable media, and methods for generating training data for training an ordinal, regression-based classifier, the method including grouping client data based on client keys associated with the client data, pairwise matching records in the client data to generate feature signatures and inferring a label based on client key statuses for the pairwise-matched records, and building a training dataset from the inferred labels and feature signatures, the training dataset used to train the classifier.

公开(公告)号：US20230126932A1

公开(公告)日：2023-04-27

申请号：US17511780

申请日：2021-10-27

Applicant: AMPERITY, INC.

Inventor： Yan YAN ,  Christopher James CHAPO ,  Joseph CHRISTIANSON ,  Joyce GORDON ,  Andrew LIM ,  Nicholas RESNICK

IPC: G06Q30/02 ,  G06F7/02

Abstract: The example embodiments are directed toward improvements in predicting an ideal audience size. In an embodiment, a method is disclosed comprising receiving a set of users associated with an object attribute; selecting samples from the set of users; computing hit rates for the samples, a respective hit rate in the hit rates computed by calculating a total number of users in a respective sample associated with an interaction associated with the object attribute; and selecting a recommended sample from the samples, the recommended sample comprising a sample having an associated hit rate that meets a preconfigured hit rate threshold.