Capital One Services, LLC

Noriaki Tatsumi

Austin Grant Walters

Jeremy Edward Goodsitt

Anh Truong

Fardin Abdi Taghi Abad

Mark Watson

Kate Key

Vincent Pham

Reza Farivar

Michael Walters

Christian Bartram

Connor Cason

Kenneth Taylor

A cloud computing system can be configured to generate a synthetic data stream that tracks a reference data stream. A model optimizer of the cloud computing system can receive, from an interface of the cloud computing system, a synthetic data stream request indicating a reference data stream. A dataset generator of the cloud computing system can generate a synthetic data stream that tracks the reference data stream by repeatedly swapping data models of the reference data stream. One such repeat can include retrieving, by the dataset generator from a model storage, a current data model of the reference data stream and generating a new data model of the reference data stream. The model optimizer can store the new data model in the model storage. The dataset generator can generate a synthetic data stream using the current data model of the reference data stream.

Systems and methods for data stream simulation

In some implementations, a system may monitor session data associated with a first module and a second module of a platform. The system may determine a rate of communication between the first module and the second module based on the session data. The system may determine, using an optimization model, a co-location score associated with the first module and the second module based on the rate of communication, wherein the co-location score indicates an impact of co-location of the first module and the second module. The system may determine that the co-location score satisfies a co-location score threshold associated with an improvement to an operation of the platform. The system may perform an action associated with co-locating the first module and the second module.

Dynamic management of locations of modules of a platform hosted by a distributed system

A cloud computing system can be configured to generate data models. A model optimizer of the cloud computing system can provision computing resources of the cloud computing system with a data model. A dataset generator of the cloud computing system can generate a synthetic dataset for training the data model. The computing resources can train the data model using the synthetic dataset. The model optimizer can store the data model and metadata of the data model in a model storage. The cloud computing system can receive production data from a data source by a production instance of the cloud computing system using a common file system. The production data can be processed using the data model by the production instance. The computing resources, the dataset generator, and the model optimizer can be hosted by separate virtual computing instances of the cloud computing system.

Systems and methods for synthetic data generation

Various embodiments are generally directed to techniques to reduce inputs of a machine learning model (MLM) and increase path efficiency as a result. A method for reducing an MLM includes: receiving a machine learning (ML) dataset, partitioning the ML dataset into a first dataset, a second dataset, a third dataset, and a fourth dataset, training, validating, and testing the MLM using one or more of the first dataset, the second dataset, and the third dataset, after testing the MLM, automatically ranking an importance associated with each input of the MLM using the fourth dataset, and reducing a plurality of inputs of the MLM based on the automatic ranking.

Automatically reducing machine learning model inputs

A method for microservices architecture optimization is disclosed. The method includes receiving a first application request message at a gateway service, and generating a first client request message, by the gateway service, based on the first application request message. The first client request message may have a customized header comprising an identification of the gateway service. The method may include forwarding the first client request message from the gateway service to a first service of a plurality of services, and the first service updating the customized header of the first client request message to add an identification of the first service. The method includes generating, at the first service, a first client response message comprising the customized header, and generating, at the gateway service, a deployment scheme for a subset of services of the plurality of services based on the customized header of the first client response message. The deployment scheme may utilize an application specific criteria.

Automated identification of service boundaries based on system focus optimization using distributed tracing data

Automated systems and methods for optimizing a model are disclosed. For example, in an embodiment, a method for optimizing a model may comprise receiving a data input that includes a desired outcome and an input dataset identifier. The method may include retrieving an input dataset based on the identifier and receiving an input model based on the desired outcome. The method may also comprise using a data synthesis model to create a synthetic dataset based on the input dataset and a similarity metric. The method may also comprise debugging the input model using synthetic dataset to create a debugged model. The method may also comprise selecting an actual dataset based on the input dataset and the desired outcome. In some aspects, the method may comprise optimizing the debugged model using the actual dataset and storing the optimized model.

Growing community of inventors

Silver Spring, MD, United States of America

Noriaki Tatsumi