18649545. METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS simplified abstract (State Farm Mutual Automobile Insurance Company)

METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS

Organization Name

State Farm Mutual Automobile Insurance Company

Inventor(s)

Jeffrey S. Myers of Normal IL (US)

Kenneth J. Sanchez of San Francisco CA (US)

Michael L. Bernico of Bloomington IL (US)

METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS - A simplified explanation of the abstract

This abstract first appeared for US patent application 18649545 titled 'METHOD OF CONTROLLING FOR UNDESIRED FACTORS IN MACHINE LEARNING MODELS

The abstract describes a method of training and using a machine learning model that excludes undesired factors such as age, sex, ethnicity, and race when analyzing new data, such as images of insurance applicants, to determine appropriate insurance premiums.

• The model is trained to correlate aspects of an applicant's appearance with personal and health-related characteristics, excluding undesired factors.
• The trained model analyzes new data, like images of applicants, without considering the excluded factors to suggest appropriate insurance premiums based on desired factors only.

Potential Applications: This technology can be applied in the insurance industry for underwriting processes to determine fair and unbiased insurance premiums based on relevant factors only.

Problems Solved: This technology addresses the issue of bias in insurance underwriting by excluding undesired factors such as age, sex, ethnicity, and race from the analysis.

Benefits: The technology ensures fair and unbiased insurance premium calculations by focusing solely on relevant factors related to the applicant's appearance and health.

Commercial Applications: Title: "Unbiased Insurance Underwriting Technology" This technology can be used by insurance companies to streamline the underwriting process, improve accuracy in premium calculations, and enhance fairness in insurance pricing.

Prior Art: Readers can explore prior research on machine learning models in insurance underwriting and bias mitigation strategies to understand the existing knowledge in this field.

Frequently Updated Research: Stay updated on advancements in machine learning algorithms for bias mitigation in insurance underwriting and related studies on fairness in artificial intelligence applications.

Questions about the Technology: 1. How does this technology ensure fairness in insurance premium calculations? 2. What are the key factors considered by the machine learning model when analyzing new data for insurance underwriting?

Original Abstract Submitted

A method of training and using a machine learning model that controls for consideration of undesired factors which might otherwise be considered by the trained model during its subsequent analysis of new data. For example, the model may be a neural network trained on a set of training images to evaluate an insurance applicant based upon an image or audio data of the insurance applicant as part of an underwriting process to determine an appropriate life or health insurance premium. The model is trained to probabilistically correlate an aspect of the applicant's appearance with a personal and/or health-related characteristic. Any undesired factors, such as age, sex, ethnicity, and/or race, are identified for exclusion. The trained model receives the image (e.g., a “selfie”) of the insurance applicant, analyzes the image without considering the identified undesired factors, and suggests the appropriate insurance premium based only on the remaining desired factors.