Abstract
We present ELQ (Efficiently Learning Quantum systems), an open-source framework for designing optimal quantum control schemes that achieve high-fidelity gate operations. Our method leverages automatic differentiation and GPU acceleration in JAX to efficiently compute gradients through realistic experimental constraints, including bandwidth limitations, non-linear transmission functions, and environmental noise. We demonstrate ELQ's capabilities by optimizing single-, simultaneous single-, and two-qubit gates for hole spins in silicon quantum dots. By incorporating numerically sampled noise trajectories, we develop control pulses that are robust to both high-frequency noise from ensembles of two-state fluctuators (1/f noise) and quasi-static errors from charge and magnetic field variations. Our results show that pulses optimized for simultaneous single-qubit operations significantly outperform those designed for sequential operation, highlighting the importance of considering cross-talk and unwanted coupling effects. The ELQ codebase enables the broader quantum computing community to design and optimize high-fidelity quantum gates while accounting for device-specific constraints and noise environments.
*This work was supported by the Royal Society, the EPSRC National Quantum Technology Hub in Networked Quantum Information Technology (EP/M013243/1), Quantum Technology Capital (EP/N014995/1), EPSRC Platform Grant (EP/R029229/1), the European Research Council (Grant agreement 818751), Graphcore, the Swiss NSF Project 179024, the Swiss Nanoscience Institute, the NCCR QSIT, the NCCR SPIN, and the EU H2020 European Microkelvin Platform EMP Grant No. 824109. This publication was also made possible through support from Templeton World Charity Foundation and John Templeton Foundation. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Templeton Foundations.
