Artificial Intelligence and Machine Learning in QubiC

Artificial intelligence (AI) and machine learning (ML) are pivotal for the advancement of quantum computing, offering new paradigms for control and optimization. Qubit Control (QubiC) is a field-programmable gate array (FPGA) based open-source full-stack control system designed specifically for quantum computing. Within the QubiC system, we explore the application of AI and ML to enhance quantum control capabilities.

We implement a neural network model on FPGA for real-time quantum state discrimination. Additionally, we create an online learning framework on FPGA that updates the state discrimination model in situ, effectively countering qubit drift. We also develop a large language model (LLM) AI agent for automatic qubit calibration. Furthermore, we use LLM to optimize hardware description language (HDL) code for quantum applications and to generate scripts for instrument manipulation in the lab.

The AI/ML advancements in QubiC will provide a versatile toolbox, paving the way for significant progress in quantum control.