Visa International Service Association

University of Texas System

Dan Wang

Yiwei Cai

Harishkumar Sundarji Majithiya

Roshni Ann Samuel

Claudia Barcenas

Youxing Qu

Himanshu Chauhan

Susan Finnegan

Xi Kan

Yu Gu

Peter A Walker

Chiranjeet Chetia

Hongqin Song

Shi Cao

Peng Wu

Sheng Wang

Qingguo Chen

Shuo Wang

Fengyi Gao

Systems, methods, and computer program products are provided for generating and improving multitask learning models. An example method includes determining a first accuracy metric based on at least two machine learning models performing a plurality of tasks, receiving a multitask learning model including at least one shared layer and a plurality of task-specific layers, determining a second accuracy metric based on the multitask learning model having a first number of shared layers, determining a third accuracy metric based on the multitask learning model having a second number of shared layers, comparing the accuracy metrics, repeating until at least one termination condition is satisfied, and determining a target number of shared layers for the multitask learning model based on at least one of the second accuracy metric, the third accuracy metric, the first number of shared layers, the second number of shared layers, or any combination thereof.

System, method, and computer program product for generating and improving multitask learning models

Methods, systems, and computer program products are provided for cleaning noisy data from unlabeled datasets using autoencoders. A method includes receiving training data including noisy samples and other samples. An autoencoder network is trained based on the training data to increase a first metric based on the noisy samples and to reduce a second metric based on the other samples. Unlabeled data including unlabeled samples is received. A plurality of third outputs is generated by the autoencoder network based on the plurality of unlabeled samples. For each respective unlabeled sample, a respective third metric is determined based on the respective unlabeled sample and a respective third output, and whether to label the respective unlabeled sample as noisy or clean is determined based on the respective third metric and a threshold. Each respective unlabeled sample determined to be labeled as noisy is cleaned.

System, method, and computer program product for cleaning noisy data from unlabeled datasets using autoencoders

A computer-implemented method includes: receiving an inquiry request message identifying a first payment transaction having a plurality of transaction parameters and a risk score, where the risk score is generated by a machine-learning model based on the plurality of transaction parameters; for each transaction parameter of the plurality of transaction parameters, perturbing a value of the transaction parameter and re-analyzing the first payment transaction with the machine-learning model to generate a perturbed risk score based on the perturbed transaction parameter; determining at least one impact parameter from the plurality of transaction parameters by comparing the perturbed risk scores generated for each of the plurality of transaction parameters; and generating an inquiry response message based on the at least one impact parameter.

System, method, and computer program product for interpreting black box models by perturbing transaction parameters

Methods for generating fraud detection rules based on transaction data may include receiving historical transaction data, associating tags with each transaction, generating decision trees having root nodes and child nodes operably connected to the respective root nodes, determining at least one primary rule and at least one set of secondary rules associated with the at least one primary rule based on relationships between features of the transactions, assigning primary rules and sets of secondary rules to the at least one decision tree to populate the tree, extracting a plurality of rule sets including at least one primary rule and one or more secondary rules, determining an ordering of the plurality of rule sets; and determining a subset of rule sets from the ordered plurality of rule sets against which subsequently received transactions are compared against to determine if the subsequent transactions are fraudulent.

System, method, and computer program product for determining fraud rules

Embodiments of the invention are directed to systems and methods for utilizing a cache to store historical transaction data. A predictive model may be trained to identify particular identifiers associated with historical data that is likely to be utilized on a particular date and/or within a particular time period. The historical data corresponding to these identifiers may be stored in a cache of the processing computer. Subsequently, an authorization request message may be received that includes an identifier. The processing computer may utilize the identifier to retrieve historical transaction data from the cache. The retrieved data may be utilized to perform any suitable operation. By predicting the data that will be needed to perform these operations, and preemptively store such data in a cache, the latency associated with subsequent processing may be reduced and the performance of the system as a whole improved.

Techniques for utilizing a predictive model to cache processing data

Growing community of inventors

Austin, TX, United States of America

Dan Wang