Teaching atoms to compute: advancing neutral atom quantum technology
Madhav Mohan defended his PhD thesis at the Department of Applied Physics and Science Education on October 21.
Madhav Mohan has developed innovative strategies to make quantum computers based on neutral atoms faster, more reliable, and more efficient. His work addresses key challenges in the noisy intermediate-scale quantum (NISQ) era, where hardware imperfections and environmental noise limit the performance of quantum devices.
Fighting noise with smarter controls
Neutral atom quantum computers trap individual atoms using laser beams and use their electronic states as qubits. These systems promise excellent scalability, but their operations must be extremely precise. Mohan created a detailed model to predict error sources, from atomic decay to laser fluctuations, and designed specialized laser pulses that remain robust even under imperfect conditions. These pulses achieve over 99.9% accuracy, significantly improving the reliability of basic quantum operations.
Speeding up gates with AI
Quantum algorithms rely on sequences of gates, some of which require fine-tuned control. Traditionally, generating accurate pulses for all possible gate settings is slow and complex. Mohan introduced an AI-driven approach using neural networks to automatically design entire families of control pulses. These AI-generated gates are not only precise but also up to 4.6 times faster than conventional methods, reducing computation time and minimizing the impact of noise.
Co-designing algorithms and hardware
Beyond improving individual operations, Mohan explored how to customize algorithms for neutral atom hardware, focusing on quantum chemistry applications. By co-designing algorithms and hardware-native gate sets, he drastically reduced runtime and resource requirements for simulating molecules, a critical step toward applications in drug discovery and materials science.
A practical path to near-term quantum computing
Mohan鈥檚 research demonstrates that progress in quantum computing doesn鈥檛 have to wait for flawless hardware. By combining robust control techniques, AI optimization, and hardware-aware algorithms, his work provides a multi-level toolkit to boost the performance of today鈥檚 quantum devices. These advances bring us closer to realizing quantum computers capable of solving real-world problems in science and industry.
-
PhD student
Madhav Mohan, Department of Applied Physics and Science Education
Read more -
Supervisors
Servaas Kokkelmans and Edgar Vredenbregt