Speeding up quantum perceptron via shortcuts to adiabaticity

Area of Science:

• Quantum computing
• Quantum machine learning
• Artificial intelligence

Background:

• The quantum perceptron is a foundational element in quantum machine learning.
• Quantum machine learning integrates quantum computing principles like superposition and entanglement with classical machine learning.
• Existing quantum perceptron protocols often rely on quasi-adiabatic methods.

Purpose of the Study:

• To propose a speed-up quantum perceptron.
• To enhance perceptron performance and robustness using inverse engineering techniques.
• To leverage shortcuts to adiabaticity for rapid quantum processing.

Main Methods:

• Utilizing shortcuts to adiabaticity principles.
• Implementing inverse engineering for control fields.
• Designing a quantum perceptron with a sigmoid activation function.
• Analyzing performance against quasi-adiabatic protocols.

Main Results:

• The proposed quantum perceptron demonstrates significantly faster operation compared to quasi-adiabatic approaches.
• The speed-up is achieved through an inversely engineered control field.
• The model exhibits enhanced robustness against control imperfections.
• A rapid nonlinear response with a sigmoid activation function is achieved.

Conclusions:

• The inversely engineered speed-up quantum perceptron offers superior performance and resilience.
• This work advances the development of efficient quantum machine learning algorithms.
• The findings pave the way for more practical quantum artificial intelligence applications.