20230419096. SYSTEMS AND METHODS FOR CONCEALING UNINTERESTED ATTRIBUTES IN MULTI-ATTRIBUTE DATA USING GENERATIVE ADVERSARIAL NETWORKS simplified abstract (JPMorgan Chase Bank, N.A.)

SYSTEMS AND METHODS FOR CONCEALING UNINTERESTED ATTRIBUTES IN MULTI-ATTRIBUTE DATA USING GENERATIVE ADVERSARIAL NETWORKS

Organization Name

JPMorgan Chase Bank, N.A.

Inventor(s)

Richard Chen of Baldwin Place NY (US)

Marco Pistoia of Amawalk NY (US)

Shaohan Hu of Yorktown Heights NY (US)

Bill Moriarty of West Chester PA (US)

Hargun Kalsi of Monmouth Junction NJ (US)

SYSTEMS AND METHODS FOR CONCEALING UNINTERESTED ATTRIBUTES IN MULTI-ATTRIBUTE DATA USING GENERATIVE ADVERSARIAL NETWORKS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20230419096 titled 'SYSTEMS AND METHODS FOR CONCEALING UNINTERESTED ATTRIBUTES IN MULTI-ATTRIBUTE DATA USING GENERATIVE ADVERSARIAL NETWORKS

Simplified Explanation

The abstract describes a patent application for systems and methods that use generative adversarial networks to conceal uninterested attributes in multi-attribute data. Here is a simplified explanation of the abstract:

• The patent application describes a method for concealing uninterested attributes in multi-attribute data using generative adversarial networks.
• The method involves pretraining a variational autoencoder to separate each attribute in the data and a decoder to reconstruct the data.
• Additional data sets are received, along with an identification of the uninterested attribute to conceal and the interested attribute to retain.
• A multi-layer perceptron is trained using the variational encoder, decoder, additional data sets, uninterested attribute, and interested attribute.
• The method also involves receiving multi-attribute data for processing and using the encoder, multi-level perceptron, decoder, and additional data sets to process the data.
• The result is multi-attribute data with the uninterested attribute concealed and the interested attributes retained.

Potential Applications

This technology has potential applications in various fields, including:

• Data privacy: Concealing uninterested attributes can help protect sensitive information in multi-attribute data.
• Data analysis: By concealing uninterested attributes, analysts can focus on the relevant attributes and reduce noise in the data.
• Machine learning: This technology can be used to preprocess data for machine learning models, improving their performance and efficiency.

Problems Solved

The technology described in this patent application solves several problems, including:

• Privacy concerns: By concealing uninterested attributes, sensitive information can be protected, addressing privacy concerns.
• Data noise: Removing uninterested attributes reduces noise in the data, making it easier to analyze and process.
• Data efficiency: By focusing on the interested attributes, the technology improves the efficiency of data analysis and machine learning algorithms.

Benefits

The use of generative adversarial networks to conceal uninterested attributes in multi-attribute data offers several benefits, including:

• Enhanced privacy: Sensitive information can be concealed, ensuring privacy and compliance with data protection regulations.
• Improved data analysis: By removing uninterested attributes, analysts can focus on the relevant information, leading to more accurate and meaningful insights.
• Efficient machine learning: Preprocessing data to conceal uninterested attributes can improve the efficiency and performance of machine learning models.

Original Abstract Submitted

systems and methods for concealing uninterested attributes in multi-attribute data using generative adversarial networks are disclosed. in one embodiment, a method may include: an attribute concealing computer program receiving multi-attribute training data from a data source; pretraining a variational autoencoder to separate each attribute in the multi-attribute training data into a space; pretraining a decoder to reconstruct data from the spaces; receiving a plurality of additional data sets; receiving an identification of an uninterested attribute to conceal and an interested attribute to retain; training a multi-layer perceptron using the variational encoder, the decoder, the additional data sets, the uninterested attribute, and the interested attribute; receiving multi-attribute data for processing; and processing the multi-attribute data using the encoder, the multi-level perceptron, the decoder, and the additional data sets, wherein the processing results in the multi-attribute data with the uninterested attribute concealed and the interested attributes retained.