Delft University Team Uses Machine Learning to Enhance Quantum Computing Performance

Researchers from the Department of Quantum & Computer Engineering at Delft University of Technology have developed a novel approach to Quantum Gate Set Tomography (QGST) using machine learning techniques. The team, including King Yiu Yu, Aritra Sarkar, Ryoichi Ishihara, and Sebastian Feld, used a transformer neural network model to address the computational complexity of modeling quantum systems. The research marks a significant step in applying deep neural networks to the complex problem of QGST, demonstrating the potential of machine learning in tackling nonlinear tomography challenges in quantum computing.

Quantum Computing: A New Approach to Gate Set Tomography

Quantum computing, a promising frontier in high-performance computing, blends quantum information theory with practical applications to overcome the limitations of classical computation. This study investigates the challenges of manufacturing high-fidelity and scalable quantum processors. Quantum gate set tomography (QGST) is a critical method for characterizing quantum processors and understanding their operational capabilities and limitations.

Machine Learning for Quantum Gate Set Tomography

This research introduces Ml4Qgst, a novel approach to QGST by integrating machine learning techniques, specifically utilizing a transformer neural network model. Adapting the transformer model for QGST addresses the computational complexity of modeling quantum systems. Advanced training strategies, including data grouping and curriculum learning, are employed to enhance model performance, demonstrating significant congruence with ground-truth values.

Quantum Computation and Gate Set Tomography

Quantum computation is an emerging paradigm that has captured the attention of theoretical physicists and computer scientists, as well as stakeholders in high-performance computing. Quantum algorithms can solve problems in specific complexity classes that are asymptotically intractable in all implementations of classical computation. A critical criterion is the characterization of the quantum processor, which helps in understanding the fabrication defects and the computing capabilities of these systems.

Transformer Model for Quantum Tomography

Various machine learning techniques are used to address the computational cost in tomography. However, these techniques have not yet been utilized for quantum gate set tomography. Our contribution presented in this work is three-fold. Firstly, we develop a first-of-its-kind machine-learning model