Research Experience
Centre for Quantum Technologies, National University of Singapore (NUS)
July 2020 – May 2021
Quantum Optimal Control using Deep Reinforcement Learning
- Designed and implemented a Deep Reinforcement Learning (DRL) environment using OpenAI Gym to generate control pulses for driving quantum state transitions.
- Formulated reward functions and termination criteria to balance fidelity achievement with control amplitude constraints.
- Tested the SAC algorithm on qubit and 3-level Transmon systems, analyzing learning performance across 8 different experimental configurations.
- Utilized Python libraries including
stable-baselines3,qutip, andnumpyfor simulation and data analysis.
IQ Mixer Calibration and Automation
- Developed a systematic calibration procedure to minimize Local Oscillator (LO) leakage and correct IQ imbalance in quantum measurement systems.
- Implemented a general 2D minimization algorithm to optimize DC offsets, gain, and phase parameters across multiple analog output channels.
- Automated the tuning process using Python, integrating control of Quantum Machines (QM) hardware and spectrum analyzers.
- Documented the full setup, calibration steps, and results, improving reproducibility and efficiency for experimentalists.