Optimizing Quantum Control Pulses with Gaussian Process Priors: The Spectral Way.

Area of Science:

• Quantum engineering
• Laser physics
• Computational chemistry

Background:

• Laser pulse shaping is crucial for controlling quantum systems.
• Existing methods can be computationally intensive and sensitive to experimental noise.
• Efficient and robust pulse shaping is needed for practical quantum applications.

Purpose of the Study:

• To introduce a novel methodology, Gaussian Process Prior Optimization for Pulse Shaping (GPPOPS), for efficient laser pulse shaping.
• To identify experimentally implementable laser pulse shapes that optimize specific tasks, such as maximizing molecular transitions.
• To demonstrate the robustness and versatility of the GPPOPS approach.

Main Methods:

• Development and application of the GPPOPS methodology.
• Utilizing a surrogate model of the control landscape for optimization.
• Testing the method on the AlH+ molecule to optimize vibronic transitions.

Main Results:

• GPPOPS successfully identified optimal laser pulse shapes for maximizing vibronic transitions in AlH+.
• The derived pulse shapes are implementable with current laser technology.
• The control capabilities of the optimized pulses demonstrated robustness against noise.

Conclusions:

• GPPOPS offers a versatile, efficient, and experimentally practical approach to pulse shaping engineering.
• The method's noise robustness distinguishes it from other numerical techniques.
• GPPOPS has the potential to significantly reduce experimental effort and drive progress in quantum engineering.