公开(公告)号：US20240386327A1

公开(公告)日：2024-11-21

申请号：US18667221

申请日：2024-05-17

Applicant: Visa International Service Association

Inventor： Yan Zheng ,  Prince Osei Aboagye ,  Michael Yeh ,  Junpeng Wang ,  Huiyuan Chen ,  Xin Dai ,  Liang Wang ,  Wei Zhang

IPC: G06N20/00

Abstract: Methods, systems, and computer program products are provided for embedding learning to provide uniformity and orthogonality of embeddings. A method may include receiving a dataset that includes a plurality of data points including a first plurality of data points having a first classification and a second plurality of data points having a second classification, generating a first normalized class mean vector of the first plurality of data instances having the first classification, generating a second normalized class mean vector of the second plurality of data instances having the second classification, performing a class rectification operation on the first plurality of data instances having the first classification and the second plurality of data instances having a second classification, and generating embeddings of the dataset based on original embedding space projections of the dataset.

公开(公告)号：US20220414662A1

公开(公告)日：2022-12-29

申请号：US17358575

申请日：2021-06-25

Applicant: Visa International Service Association

Inventor： Shi Cao ,  Chiranjeet Chetia ,  Liang Wang ,  Junpeng Wang ,  Morvarid Jamalian

IPC: G06Q20/40 ,  G06Q30/02 ,  G06N20/00

Abstract: A method for detecting collusive transaction fraud includes: generating a merchant baseline including a transaction data baseline and a time series baseline; extracting time series data of the first merchant system; generating a first score and second score with a deep learning model; generating a first merchant risk score of the first merchant system based on the first and second scores; in response to determining that the first merchant risk score satisfies the threshold, determining a plurality of related entities related to the first merchant system; and classifying the first merchant system and at least one related entity of the plurality of related entities in a first group risk class based on at least one risk score of the at least one related entity.

公开(公告)号：US11487997B2

公开(公告)日：2022-11-01

申请号：US16593731

申请日：2019-10-04

Applicant: Visa International Service Association

Inventor： Liang Gou ,  Junpeng Wang ,  Wei Zhang ,  Hao Yang

IPC: G06K9/62 ,  G06N3/04 ,  G06N3/08 ,  G06F16/904 ,  G06F17/18 ,  G06F17/16

Abstract: A method for local approximation of a predictive model may include receiving unclassified data associated with a plurality of unclassified data items. The unclassified data may be classified based on a first predictive model to generate classified data. A first data item may be selected from the classified data. A plurality of generated data items associated with the first data item may be generated using a generative model. The plurality of generated data items may be classified based on the first predictive model to generate classified generated data. A second predictive model may be trained with the classified generated data. A system and computer program product are also disclosed.

公开(公告)号：US20250111011A1

公开(公告)日：2025-04-03

申请号：US18374799

申请日：2023-09-29

Applicant: Visa International Service Association

Inventor： Junpeng Wang ,  Minghua Xu ,  Shubham Jain ,  Yan Zheng ,  Michael Yeh ,  Liang Wang ,  Wei Zhang

IPC: G06F17/18

Abstract: Methods, systems, and computer program products are provided for coordinated analysis of output scores and input features of machine learning models in different environments. An example method includes receiving a plurality of first data records and a plurality of second data records. A first plot is generated based on a first score generated by a machine learning model for each first data record and a second score generated by the machine learning model for each second data record. The first plot is displayed. A plurality of second plots associated with at least a subset of the plurality of features are generated. Each respective second plot is generated based on a respective first field associated with a respective feature from the first data records and a respective second field associated with the respective feature from the second data records. The second plots are displayed.

公开(公告)号：US20240289613A1

公开(公告)日：2024-08-29

申请号：US18656024

申请日：2024-05-06

Applicant: Visa International Service Association

Inventor： Haoyu Li ,  Junpeng Wang ,  Liang Wang ,  Yan Zheng ,  Wei Zhang

IPC: G06N3/08 ,  G06N3/0455

CPC classification number: G06N3/08 ,  G06N3/0455

Abstract: A method, system, and computer program product is provided for embedding compression and reconstruction. The method includes receiving embedding vector data comprising a plurality of embedding vectors. A beta-variational autoencoder is trained based on the embedding vector data and a loss equation. The method includes determining a respective entropy of a respective mean and a respective variance of each respective dimension of a plurality of dimensions. A first subset of the plurality of dimensions is determined based on the respective entropy of the respective mean and the respective variance for each respective dimension of the plurality of dimensions. A second subset of the plurality of dimensions is discarded based on the respective entropy of the respective mean and the respective variance for each respective dimension of the plurality of dimensions. The method includes generating a compressed representation of the embedding vector data based on the first subset of dimensions.

公开(公告)号：US20240127035A1

公开(公告)日：2024-04-18

申请号：US18275598

申请日：2022-02-01

Applicant: VISA INTERNATIONAL SERVICE ASSOCIATION

Inventor： Michael Yeh ,  Zhongfang Zhuang ,  Junpeng Wang ,  Yan Zheng ,  Javid Ebrahimi ,  Liang Wang ,  Wei Zhang

IPC: G06N3/0455

CPC classification number: G06N3/0455

Abstract: A method performed by a computer is disclosed. The method comprises receiving interaction data between electronic devices of a plurality of entities. The interaction data is used to form an entity interaction vector containing a number of interactions between the electronic devices of a chosen entity and an entity time series containing a plurality of metrics per unit time of the interactions. An interaction encoder of the computer can generate an interaction hidden representation of the entity interaction vector using embeddings of the plurality of entities. A temporal encoder of the computer can generate a temporal hidden representation of the entity time series. The interaction hidden representation and the temporal hidden representation can be used to generate a predicted scale and a shape estimation of a target interaction metric. The computer can then generate an estimated interaction metric of a time period using the predicted scale and the shape estimation.

公开(公告)号：US20230186078A1

公开(公告)日：2023-06-15

申请号：US17912070

申请日：2021-04-30

Applicant: Visa International Service Association

Inventor： Junpeng Wang ,  Wei Zhang ,  Hao Yang ,  Michael Yeh ,  Liang Wang

IPC: G06N3/08 ,  G06T11/20 ,  G06Q20/40

CPC classification number: G06N3/08 ,  G06T11/206 ,  G06Q20/4016 ,  G06T2200/24

Abstract: A method for evaluating a RNN-based deep learning model includes: receiving model data generated by the RNN-based model, the model data including a plurality of events associated with a plurality of states; generating a first GUI based on the events and states including a chart visually representing a timeline for the events in relation to a parameter value; generating a second GUI including a point chart visually representing a two-dimensional projection of the multi-dimensional intermediate data, each point of the point chart representing a time step and an event from the time step, based on multi-dimensional intermediate data between transformations in the model that connect a state to an event; and perturbing the environment at a time step based on user interaction with at least one of the first and second GUIs.

公开(公告)号：US12118462B2

公开(公告)日：2024-10-15

申请号：US17148984

申请日：2021-01-14

Applicant: Visa International Service Association

Inventor： Zhongfang Zhuang ,  Michael Yeh ,  Liang Wang ,  Wei Zhang ,  Junpeng Wang

IPC: G06N3/08 ,  G06N3/044 ,  G06N3/045

CPC classification number: G06N3/08 ,  G06N3/044 ,  G06N3/045

Abstract: Described are a system, method, and computer program product for multivariate event prediction using multi-stream recurrent neural networks. The method includes receiving event data from a sample time period and generating feature vectors for each subperiod of each day. The method also includes providing the feature vectors as inputs to a set of first recurrent neural network (RNN) models and generating first outputs for each RNN node. The method further includes merging the first outputs for each same subperiod to form aggregated time-series layers. The method further includes providing the aggregated time-series layers as an input to a second RNN model and generating final outputs for each RNN node of the second RNN model.

公开(公告)号：US20240273095A1

公开(公告)日：2024-08-15

申请号：US18567717

申请日：2022-06-01

Applicant: Visa International Service Association

Inventor： Michael Yeh ,  Yan Zheng ,  Junpeng Wang ,  Wei Zhang ,  Zhongfang Zhuang

IPC: G06F16/2453 ,  G06F16/2458

CPC classification number: G06F16/24537 ,  G06F16/2465 ,  G06F16/2477

Abstract: A method is disclosed. The method comprises determining a time series, a subsequence length. The length of the time series may then be determined, and an initial matrix profile may then be computed. The method may then form a processed matrix profile for a first subsequence of the subsequence length by applying the first subsequence to the initial matrix profile. A second subsequence may then be determined from the processed matrix profile. The method may then include comparing the second subsequence to other subsequences in a dictionary and adding it to the dictionary. The subsequences in the dictionary may be used to generate a plurality of subsequence matrix profiles. The method may then include forming an approximate matrix profile using the plurality of subsequence matrix profiles and then determining one or more anomalies in the time series or another time series using the approximate matrix profile.

公开(公告)号：US20240160854A1

公开(公告)日：2024-05-16

申请号：US18280792

申请日：2022-03-30

Applicant: Visa International Service Association

Inventor： Sunipa Dev ,  Yan Zheng ,  Michael Yeh ,  Junpeng Wang ,  Wei Zhang ,  Archit Rathore

IPC: G06F40/40

CPC classification number: G06F40/40

Abstract: Described are a system, method, and computer program product for debiasing embedding vectors of machine learning models. The method includes receiving embedding vectors and generating two clusters thereof. The method includes determining a first mean vector of the first cluster and a second mean vector of the second cluster. The method includes determining a bias associated with each of a plurality of first candidate vectors and replacing the first mean vector with a first candidate vector based on the bias. The method includes determining a bias associated with each of a plurality of second candidate vectors and replacing the second mean vector with a second candidate vector based on the bias. The method includes repeatedly replacing the first and second mean vectors until an extremum of the bias score is reached, and debiasing the embedding vectors by linear projection using a direction defined by the first and second mean vectors.