Abstract:
Quantum computing has emerged as a powerful paradigm for addressing difficult computational tasks that classical systems cannot handle. Nevertheless, quantum algorithms are sensitive to the optimisation and design of quantum circuits, especially in noisy intermediate-scale quantum settings. This paper provides a specialised systematic literature review on the use of artificial intelligence methods in the optimisation of quantum circuits. After completing a systematic search and screening process, thirteen peer-reviewed publications from 2017 to 2026 have been selected for analysis, including machine learning, reinforcement learning, deep learning, and evolutionary algorithms. The studies discussed above show that methods based on artificial intelligence could help improve the performance of circuits by decreasing their depth, the number of gates used, and making the computations more efficient. From the above analysis, it is possible to develop an AI-based framework where all the key AI-based methodologies and optimisation goals that emerged from the analysed literature have been incorporated. Another trend in the area of optimisation process development, which is mentioned in the review, is the increasing significance of hybrid quantum-classical methods and quantum machine learning. Though there have been these improvements, issues like scarcity of hardware, problems of scaling, and non-standardised evaluation frameworks still exist. This research will give a systematic overview of the current trends and gaps in the research to inform future research in this dynamic field.
