17884170. MULTITASK LEARNING BASED ON HERMITIAN OPERATORS simplified abstract (TOYOTA JIDOSHA KABUSHIKI KAISHA)

MULTITASK LEARNING BASED ON HERMITIAN OPERATORS

Organization Name

TOYOTA JIDOSHA KABUSHIKI KAISHA

Inventor(s)

[[:Category:Jens Strabo Hummelsh�j of Brisbane CA (US)|Jens Strabo Hummelsh�j of Brisbane CA (US)]][[Category:Jens Strabo Hummelsh�j of Brisbane CA (US)]]

MULTITASK LEARNING BASED ON HERMITIAN OPERATORS - A simplified explanation of the abstract

This abstract first appeared for US patent application 17884170 titled 'MULTITASK LEARNING BASED ON HERMITIAN OPERATORS

Simplified Explanation

The abstract describes a method for multitask learning using Hermitian operators in machine learning to predict properties of materials.

• Training a multitask machine learning model to map input representations of materials to complex wave function state vectors.
• Inferring observable property matrices for each observable property of the material using the trained model.
• Converting observable property matrices into complex Hermitian operators.
• Predicting target properties of the material based on the complex Hermitian operators and complex wave function state vector.

Potential Applications

This technology could be applied in various fields such as material science, chemistry, and physics for predicting properties of different materials.

Problems Solved

1. Efficient prediction of material properties. 2. Multitask learning for mapping input representations to complex wave function state vectors. 3. Converting observable property matrices into complex Hermitian operators.

Benefits

1. Improved accuracy in predicting material properties. 2. Enhanced efficiency in multitask learning. 3. Versatile application in different scientific fields.

Original Abstract Submitted

A method for multitask learning based on Hermitian operators is described. The method includes training a multitask machine learning (MTML) model to map an input representation of a material onto a complex wave function state vector. The method also includes inferring, by a trained, MTML model, observable property matrices for each observable property of the material. The method further includes converting the observable property matrices into complex Hermitian operators. The method also includes predicting target properties of the material according to the complex Hermitian operators and the complex wave function state vector.