AI-Assisted Quantum Computing for Compact Circuit Generation using CUDA-Q

Artificial Intelligence (AI) is playing an increasingly important role in advancing quantum computing. However, the adoption of AI has been hindered by the lack of standardized databases for compact quantum circuits. In this talk, I will present our work on implementing the ADAPT method using the CUDA-Q framework, enabling the fast and efficient generation of compact quantum circuits on NVIDIA GPUs. The optimized circuits will be made available to the community to facilitate the development of novel AI-driven methods to advance the field of quantum computing. These generated datasets are utilized to pre-train our ADAPT-GPT model, which is designed to produce compact circuits for new quantum problems. Preliminary results highlight the accuracy and efficiency of this approach, offering a scalable solution for quantum circuit generation.