Capital One Services, LLC

Nathan Wolfe

Purva Shanker

Tania Cruz Morales

Shannon Yogerst

Dan Lin

Ignacio Espino

Joshua S Edwards

Christopher Larson

Tarek Aziz Lahlou

Md Arafat Hossain Khan

Gang Mei

Zhuqing Zhang

Jing Zhu

Ebony Edwards

Andy Luo

Systems and methods for identifying data labels for submitting to additional data labeling routines based on embedding clusters are described. For example, the system may retrieve an unlabeled dataset, wherein the unlabeled dataset comprises a plurality of unlabeled samples. The system may generate a plurality of embeddings based on the unlabeled dataset, wherein the plurality of embeddings comprises a respective embedding for each unlabeled sample in the plurality of unlabeled samples. The system may cluster the plurality of embeddings into a plurality of clusters. The system may, based on the first cluster, determine a first labeled sample of the first labeled dataset corresponding to the first cluster and determine to submit the first labeled sample to a second data labeling routine. The system may generate, using the second data labeling routine, a second labeled sample based on the first labeled sample.

Systems and methods for identifying data labels for submitting to additional data labeling routines based on embedding clusters

In some embodiments, intent labeling may be facilitated for machine learning model training. In some embodiments, a set of natural language inputs may be obtained. Based on an action set, content related to intent labeling may be presented on a user interface. Based on an entity hierarchy and user selection of an agent-side action of the action set, a first entity subset of entities associated with the agent-side action may be presented on the user interface. Based on the entity hierarchy and user selection of a first entity from the first entity subset, a second entity subset of entities associated with the first entity may be presented on the user interface. Based on user selection of a second entity from the second entity subset, the natural language input may be associated with an ordered intent label that includes the agent-side action, the first entity, and the second entity.

Systems and methods for facilitating intent labeling for machine learning model training

In some implementations, a device may obtain a natural language input including an emoji. The device may identify one or more appearance modifiers associated with the emoji. The device may generate a token associated with the emoji that removes the one or more appearance modifiers, wherein the token is associated with multiple emojis including the emoji, and wherein the token is a modified code associated with the emoji or is associated with a cluster that is associated with the multiple emojis. The device may provide, to a natural language processing (NLP) model, the token associated with the emoji. The device may obtain, from the NLP model, an output that indicates an interpretation of the natural language input based on providing the token to the NLP model.

Emoji sanitization for natural language model processing

Systems and methods are described for maintaining bifurcated data management while labeling data for artificial intelligence model development. For example, the system may receive a first label for a first sample from a first dataset, wherein the first dataset is accessible to a first subset of a plurality of users, and wherein the first subset comprises a first attribute. The system may receive first version metadata of the first label, wherein the first version metadata comprises a proposed label for the first sample assigned by a first user. The system may determine, based on a first user input from the first user, a first grouping of source code files for storing the first version metadata, wherein the first grouping of source code files is accessible to a second subset of the plurality of users, and wherein the second subset comprises a second attribute.

Systems and methods for maintaining rights management while labeling data for artificial intelligence model development

A method includes obtaining a first dataset associated with a first set of classes and a second dataset associated with a second set of classes, where the first and second datasets share features. The method also includes creating an aggregated dataset having a quality score for the second set of classes that satisfy a first threshold. Creating the aggregated dataset includes determining a first and second set of clusters based on the first and second datasets. The method includes determining a distance based on a record of the first set of clusters and a second record of the second set of clusters in a feature space. The method further includes determining a set of analysis scores based on the distance and generating an indication for a set of records associated with the set of analysis scores based on the set of analysis scores satisfying a second threshold.

Cluster-based dataset evolution

Systems and methods for storing versioned data while labeling data for artificial intelligence model development

Systems and methods for maintaining bifurcated data management while labeling data for artificial intelligence model development

Systems, apparatuses, and methods are described for inverting neural embeddings. One or more forward neural embeddings associated with meanings, features, and/or characteristics of data samples may be generated for one or more data samples. One or more inverse neural embeddings associated with the one or more forward neural embeddings may be determined. One or more inverse feature sets for the one or more inverse neural embeddings may be generated.

Methods for exact and approximate inverses of neural embedding models

Growing community of inventors

Silver Spring, MD, United States of America

Nathan Wolfe