Capital One Services, LLC

Other

Christopher Bayan Bruss

Keegan Hines

Stephen Fletcher

Mohammad Reza Sarshogh

Lei Yu

Jakob Kressel

Samuel Sharpe

Mackenzie Sweeney

James O H Montgomery

Jason Wittenbach

Pouya Johnathon Ehsani

Antonia Gogoglou

Maximo Moyer

Methods and systems are described herein for generating updated sets of attributes that would turn a negative decision of an automated system into a positive decision. A received set of attributes associated with a negative decision of an automated system may be used to generate a latent representation of that set of attributes. A machine learning model may then be used to output a change value (an alpha value). The alpha value may represent a minimum change needed to be made to the set of attributes to change the negative decision to a positive decision. The alpha value may then be applied to the latent representation to generate an updated latent representation, which may then be decoded into an updated set of attributes that would generate a positive decision from the automated system.

Automatic generation of attribute sets for counterfactual explanations

Methods and systems are for using machine learning models to locate information in an organizational graph. A search system may use techniques described herein to determine relevant data (e.g., organizational knowledge) to retrieve from a knowledge graph for input to a machine learning model. The search system may retrieve more relevant data from the knowledge graph through the use of time data that may enable the search system to avoid outdated information. The search system may also limit the data that may be used in determining an answer to a query. By doing so, the search system may be able to answer queries more efficiently (e.g., using less computing resources, less processing power, etc.).

Machine learning for locating information in knowledge graphs

Methods and systems are described herein for generating group definition sequences for accounts (e.g., user accounts) using action sequence processing and then classifying accounts using those group definition sequences. A plurality of user account actions and corresponding time that each action was taken may be received and based on that information, a sequence of action types sometimes referred to as a time-ordered dataset of action types (e.g., based on a chronological order of the actions) may be generated. The time-ordered dataset of action types may be compared with known time-ordered sequences for a particular user group or user classification. If the time-ordered dataset of action types matches the time-ordered sequences of the particular user group, the user may be classified into that user group.

Generating user group definitions

Systems, methods, and computer program products to provide credit decisioning based on graph neural networks. A lending network graph of a plurality of loans may be received, each loan associated with a creditor and one account. A first node of the graph may be associated with a first creditor and the second node may be associated with a first account. A graph neural network may receive a respective message from each node connected to the first node, each message comprising an embedding vector reflecting a current state of the node. The graph neural network may update weights for the first node in a forward pass. The graph neural network may receive a respective message from each node connected to the second node, each message comprising the embedding vector reflecting the current state of the node. The graph neural network may update weights for the second node in a backward pass.

Credit decisioning based on graph neural networks

Methods and systems disclosed herein may quantify the content and nature of first streaming data to detect when the typical composition of the first streaming data changes. Quantifying the content and nature of the first streaming data may begin by generating a baseline representation of the content of the first streaming data as represented by a first matrix. Once generated, the first matrix may be used as a control against subsequently received data streams. In this regard, a second matrix may be generated from second streaming data and compared to the first matrix to determine the differences between the first streaming data and the second streaming data. Once a difference is determined, the difference may be compared to a threshold value and, when the difference exceeds the threshold value, an administrator may be notified and corrective action taken.

Method and system for detecting drift in text streams

A system including at least one processor; and at least one memory having stored thereon computer program code that, when executed by the at least one processor, controls the at least one processor to: receive an email addressed to a user; separate the email into a plurality of email components; analyze, using respective machine-learning techniques, each of the plurality of email components; feed the analysis of each of the plurality of email components into a stacked ensemble analyzer; and based on an output of the stacked ensemble analyzer, determine whether the email is malicious.

Systems and methods of detecting email-based attacks through machine learning

Systems, methods, and computer program products to provide neural embeddings of transaction data. A network graph of transaction data based on a plurality of transactions may be received. The network graph of transaction data may define relationships between the transactions, each transaction associated with at least a merchant and an account. A neural network may be trained based on training data comprising a plurality of positive entity pairs and a plurality of negative entity pairs. An embedding function may then encode transaction data for a first new transaction. An embeddings layer of the neural network may determine a vector for the first new transaction based on the encoded transaction data for the first new transaction. A similarity between the vectors for the transactions may be determined. The first new transaction may be determined to be related to the second transaction based on the similarity.

Neural embeddings of transaction data

Systems and methods are disclosed herein for improving machine learning of a data set. In one example, the method may include training a predictive model on an initial data set comprising labeled data, wherein the training is performed in an active learning system. The method may further include generating a set of parameters based on the training and introducing an unlabeled data set into the predictive model. According to some embodiments, the method may further include applying the set of parameters to the unlabeled data set, generating a set of predictions associated with the applied set of parameters and calculating a first uncertainty score and a second uncertainty score associated with the generated set of predictions. Moreover, the method may also include modifying the data set based on the first uncertainty score, and modifying the predictive model based on the second uncertainty score.

Systems and method for enhanced active machine learning through processing of partitioned uncertainty

Methods and systems disclosed herein may quantify a representation of a type of input an image analysis system should expect. The image analysis system may be trained on the type of input the image analysis system should expect using a first image stream. A first model of the type of input that the image analysis system should expect may be built from the first image stream. After the first model is built, a second image, or a second image stream, may be compared to the first model to determine a difference between the second image, or second image stream, and the first image stream. When the difference is greater than or equal to a threshold, a drift may be detected and steps may be taken to determine the cause of the drift.

Method and system for detecting drift in image streams

Methods and systems disclosed herein may quantify the content and nature of a first stream of text to detect when the typical composition of the first stream of text changes. Quantifying the content and nature of the first stream of text may begin by generating a baseline representation of the content of the first stream of text as represented by a first matrix. Once generated, the first matrix may be used as a control against subsequently received sequences of text. In this regard, a second matrix may be generated from a second sequence of text and compared to the first matrix to determine the differences between the first sequence of text and the second sequence of text. Once a difference is determined, the difference may be compared to a threshold value and, when the difference exceeds the threshold value, an administrator may be notified and corrective action taken.

Growing community of inventors

Washington, DC, United States of America

Christopher Bayan Bruss