Optimizing quantum algorithms for neutral atom computers
Robert de Keijzer defended his PhD thesis at the Department of Applied Physics and Science Education on December 12.
Robert de Keijzer has developed new optimization techniques for quantum computers based on neutral atoms trapped in optical tweezers. His work focuses on improving the performance of these systems in the noisy intermediate-scale quantum (NISQ) era, where hardware imperfections and noise limit computational accuracy.
Why neutral atom platforms matter
Quantum computers promise breakthroughs in fields like chemistry, materials science, and optimization. Among the various architectures, neutral atom systems stand out for their scalability and long coherence times. In Eindhoven, two such platforms are being developed, Ruby and Sapphire, the first using rubidium atoms and the second strontium, offering a promising route toward large-scale quantum computing.
However, these systems are highly sensitive to noise, which can degrade the fidelity of quantum operations. De Keijzer鈥檚 research addresses this challenge by combining mathematical modeling, optimal control theory, and algorithm design to make computations faster and more robust.
From gates to pulses: a new approach to quantum algorithms
Traditional quantum algorithms rely on gate-based methods, applying one operation at a time. De Keijzer introduced pulse-based approaches, where the entire evolution is treated as a single optimized control pulse. Using variational quantum optimal control (VQOC), he developed algorithms that leverage the strengths of neutral atom systems and outperform conventional gate-based techniques.
Building on this, he proposed Fidelity-enhanced VQOC (F-VQOC), which accounts for noise during state preparation. By applying stochastic optimal control, these methods significantly improve the accuracy of quantum computations鈥攁n essential step for practical applications in the NISQ era.
Tackling noise with advanced modeling
To understand and mitigate noise, De Keijzer used stochastic Schr枚dinger equations and It么 calculus to model how control imperfections affect qubit behavior. These models provide detailed insights into fidelity loss and guide the design of more robust algorithms. Experimental validation was performed on Ruby and supported by simulations using RySP, a digital twin of a neutral atom quantum computer under development in Eindhoven.
Toward faster and more reliable quantum computing
De Keijzer鈥檚 work demonstrates that hardware-aware optimization can dramatically improve the performance of near-term quantum devices. By tailoring algorithms to the unique capabilities of neutral atom platforms and addressing noise at its source, his research brings us closer to realizing quantum computers that can tackle real-world problems efficiently.
-
PhD student
Robert de Keijzer, Department of Applied Physics and Science Education
Read more -
Supervisors
Servaas Kokkelmans