Dell products l.p. (20240256983). SYSTEM AND METHOD FOR MANAGING LATENT BIAS IN SUPPORT VECTOR MACHINES simplified abstract
Contents
SYSTEM AND METHOD FOR MANAGING LATENT BIAS IN SUPPORT VECTOR MACHINES
Organization Name
Inventor(s)
OFIR Ezrielev of Be'er Sheva (IL)
FATEMEH Azmandian of Raynham MA (US)
SYSTEM AND METHOD FOR MANAGING LATENT BIAS IN SUPPORT VECTOR MACHINES - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240256983 titled 'SYSTEM AND METHOD FOR MANAGING LATENT BIAS IN SUPPORT VECTOR MACHINES
The abstract of the patent application describes methods, systems, and devices for providing computer-implemented services. Inference models are used to generate inferences for these services, obtained through training processes that aim to reduce bias.
- Inference models are generated and provided for computer-implemented services.
- Training processes are used to train the inference models, proactively attempting to reduce bias.
- The training process disincentivizes predictive power with respect to bias features and incentivizes predictive power for labels through debiasing terms.
Potential Applications: - This technology can be applied in various industries such as healthcare, finance, and marketing to improve decision-making processes. - It can be used in e-commerce platforms to provide personalized recommendations to users based on their preferences. - In the legal sector, this technology can assist in analyzing large amounts of data to support legal cases.
Problems Solved: - Helps reduce bias in computer-implemented services. - Improves the accuracy and reliability of inferences generated by inference models. - Enhances the overall performance of computer-implemented services by training models effectively.
Benefits: - Increased fairness and transparency in decision-making processes. - Enhanced user experience through personalized recommendations. - Improved efficiency and effectiveness of services through optimized training processes.
Commercial Applications: Title: Enhanced Decision-Making Technology for Various Industries This technology can be commercialized by offering it as a service to companies in different sectors looking to improve their decision-making processes. It can be marketed as a tool to enhance the accuracy and fairness of predictions and recommendations.
Questions about the Technology: 1. How does this technology address bias in computer-implemented services? - The technology uses training processes to proactively reduce bias in inference models, disincentivizing predictive power with respect to bias features. 2. What are the potential applications of this technology in the legal sector? - This technology can assist in analyzing large amounts of legal data to support legal cases, improving decision-making processes in the legal industry.
Original Abstract Submitted
methods, systems, and devices for providing computer implemented services are disclosed. to provide the computer implemented services, inference models may generate and provide inferences used in the computer implemented services. the inference models may be obtained through training using training data. training processes used to train the inference models may proactively to attempt to reduce the likelihood of the trained inference models exhibiting latent bias. the training process may disincentivize predictive power with respect to bias features and incentivize predictive power for labels through use of debiasing terms.
