Operators in quantum machine learning: Response properties in chemical space.

Area of Science:

• Quantum mechanics
• Quantum machine learning
• Computational chemistry

Background:

• Response operators are fundamental in quantum mechanics.
• Quantum machine learning models offer a novel approach to studying molecular properties.

Purpose of the Study:

• Investigate the application of response operators in universal quantum machine learning models.
• Evaluate the accuracy and efficiency of these models for predicting molecular response properties.

Main Methods:

• Developed a theoretical framework for using response operators in quantum machine learning.
• Conducted numerical experiments measuring potential energy response to atomic displacement and electric fields.
• Trained and tested models using varying training set sizes.

Main Results:

• Prediction errors for molecular properties, atomic forces, and dipole moments systematically decreased with increased training data.
• High accuracy was achieved even with small training datasets.
• Successfully predicted normal modes and infrared spectra for small molecules.

Conclusions:

• Response operator-based quantum machine learning provides an accurate and efficient method for predicting molecular properties.
• This approach demonstrates significant utility for computational chemistry, enabling precise predictions with minimal data.