Dell products l.p. (20240256854). SYSTEM AND METHOD FOR SELECTING MODEL TOPOLOGY simplified abstract

SYSTEM AND METHOD FOR SELECTING MODEL TOPOLOGY

Organization Name

dell products l.p.

Inventor(s)

OFIR Ezrielev of Be'er Sheva (IL)

TOMER Kushnir of Omer (IL)

FATEMEH Azmandian of Raynham MA (US)

SYSTEM AND METHOD FOR SELECTING MODEL TOPOLOGY - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240256854 titled 'SYSTEM AND METHOD FOR SELECTING MODEL TOPOLOGY

The abstract of the patent application describes methods, systems, and devices for providing computer-implemented services while managing inference models to reduce bias in the inferences provided. By using a divisional process to obtain multipath inference models, the likelihood of latent bias features being included in the inferences is reduced, thereby improving the accuracy and fairness of the computer-implemented services.

• Inference models are managed to reduce the likelihood of bias features in the inferences provided.
• Divisional process is used to obtain multipath inference models.
• Modified split training is employed to reduce mutual information shared with bias features.
• Inferences provided are less likely to include latent bias, reducing bias in computer-implemented services.
• The technology aims to improve the accuracy and fairness of the inferences provided by data processing systems.

1. 1. 1. Potential Applications:

The technology can be applied in various fields such as healthcare, finance, marketing, and social media platforms to provide more accurate and unbiased computer-implemented services.

1. 1. 1. Problems Solved:

The technology addresses the issue of bias in inference models used by data processing systems, ensuring that the inferences provided are more accurate and fair.

1. 1. 1. Benefits:

- Improved accuracy of computer-implemented services - Reduction of bias in the inferences provided - Enhanced fairness in decision-making processes

1. 1. 1. Commercial Applications:

The technology can be utilized by companies offering data processing services, AI solutions, and predictive analytics tools to improve the quality and reliability of their services, gaining a competitive edge in the market.

1. 1. 1. Questions about the Technology:
         1. 1. How does the divisional process help in obtaining multipath inference models?

The divisional process helps in creating multiple paths for the inference models to follow, reducing the likelihood of bias features being included in the inferences provided.

1. 1. 1. 1. 2. What are the potential implications of using modified split training to reduce mutual information shared with bias features?

Modified split training can help in improving the accuracy and fairness of the inferences provided by data processing systems, leading to more reliable and unbiased computer-implemented services.

Original Abstract Submitted

methods, systems, and devices for providing computer-implemented services are disclosed. to provide the computer-implemented services, inference models used by data processing systems may be managed to reduce the likelihood of the inference models provide inferences indicative of bias features. the inference models may be managed using a divisional process to obtain multipath inference models, as part of a modified split training to reduce mutual information shared with the bias feature. the inferences provided by the inference models may be less likely to include latent bias thereby reducing bias in computer-implemented services provided using the inferences.