logseq/pages/hls__MDE_Intelligence_2023_paper_18_1691442588326_0.md

file:: MDE_Intelligence_2023_paper_18_1691442588326_0.pdf file-path:: ../assets/MDE_Intelligence_2023_paper_18_1691442588326_0.pdf

• quantum information theory and linear algebra ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26a2e-c886-49f4-9685-8501654f19f4
• application of AI for model transformations, concerning endeavours such as model translation [1], generation [2], or repair [3], [4]. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26af9-5d19-4716-bc42-e98541923a8c
• program features to classes while maintaining separation of concerns, in order to obtain a high-quality software model. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26b12-ac18-4f7c-bb97-1885ec606097
• Automated Software Engineering for Quantum Computing. ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64d26b55-b55e-4cf9-b8d6-50b9689aed36
• conceptual model of quantum programs ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26b72-700e-4ab7-9bb6-d1c871410819
• synthesis of quantum programs. The latter is particularly relevant for a broad adoption of QC, because designing a suitable quantum program for a desired computational task requires extensive knowledge in quantum information theory and linear algebra ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26b96-1c88-45f4-af3b-46c1a3d79835
• no MDO approaches for the automated synthesis of quantum programs ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64d26bcf-64c1-4554-8f9e-84604bada5c9
• holistic approach that supports the definition but also to facilitate development, in parts or as a whole, by leveraging AI methods ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64d26be3-96ea-46d1-a179-4fd2ae418918
• Contributions ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64d26c65-41bc-41a1-88b1-06d68314b3a4
• ethod to conduct automated quantum program synthesis using the existing MDO engine MOMoT [7]. ls-type:: annotation hl-page:: 1 hl-color:: purple id:: 64d26c70-9036-49f8-aeed-60aa1c3f7b16
• Drawing from dedicated models for representing quantum programs and configurable MDO approaches, it would allow to further analyse the application of different AI and MDE approaches for quantum program synthesis in the future. ls-type:: annotation hl-page:: 1 hl-color:: green id:: 64d26cd6-be14-49cc-b000-98300a96bce8
• roblem-agnostic environment for model optimization. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26d7b-5fd8-40e6-8ae8-7623d6896954
• The Circuit Model ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26da3-e4bb-4048-9091-21bad154ec99
• arnessing quantum mechanical phenomen ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26da9-64e3-4abc-8744-3c55dae5b02b
• a quantum circuit comprises the application of quantum gates in an ordered manner ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26e0d-c1e8-4d81-b5c8-288ae4d61337
• gates are sequentially applied to the quantum information, which is stored on so-called qubits. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26e14-5041-4f48-95ca-04a5a7e53fbf
• Lacking the information on the implementation of Oracles prohibits the execution of a quantum circuit [35]. ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26ee7-c2ef-4d09-9d58-e6f8cb8a54a3
• he required elementary quantum gates to realize a certain functionality, is known to be a highly non-trivial task which is required for the quantum program to be executable ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 64d26f02-ce17-478d-9c24-a98a8c924004
• produced output quantum state and an expected target state ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26f1f-450d-4af0-9b07-a2702d17b4c1
• Several approaches have been explored to automate the discovery and synthesis of quantum programs, where especially reinforcement learning [15]–[17], [36]–[38] and genetic programming approaches [39]–[41] have been studied. ls-type:: annotation hl-page:: 2 hl-color:: yellow id:: 64d26f3a-a110-4ed8-b93b-15483d2d1601
• we will describe how the MOMoT framework can be applied to automatically synthesize quantum programs in terms of circuit models. ls-type:: annotation hl-page:: 2 hl-color:: purple id:: 64d26f60-6e66-4edd-8224-e8a7b75e820f
• a genetic algorithm is used to search for implementations that meet the trade-off between accuracy and computational cost present in the NISQ-era ls-type:: annotation hl-page:: 2 hl-color:: green id:: 64d26f70-b861-4bcb-89a4-23a894ee5997
• Model-Driven Optimization ls-type:: annotation hl-page:: 3 hl-color:: green id:: 64d26fcf-a2e5-4dff-8788-a05198a77c15
• Quantum Circuit Model ls-type:: annotation hl-page:: 3 hl-color:: green id:: 64d26fda-edde-4a39-b151-1ca9329afff8
• This set represents the found trade-off solutions of the multi-objective search and the Quantum Circuit Models are transformed back to the specific Q-SDK representation. ls-type:: annotation hl-page:: 3 hl-color:: purple id:: 64d2701f-fc2d-4234-99fb-fd15e5a3a0d0
• AngleParameters ls-type:: annotation hl-page:: 4 hl-color:: green id:: 64d270a5-24d3-4b94-8efa-5abc41c79951
• This paper demonstrates the feasibility of using existing model-driven search approaches for automated quantum program synthesis ls-type:: annotation hl-page:: 6 hl-color:: green id:: 64d2710c-0df5-4cb1-93c8-9c3db6b736e3
• We do not provide a performance evaluation of our proposed approach, which would comprise, among others, hyperparameter tuning, comparison of different search algorithms, and effects of additional transformation rules. ls-type:: annotation hl-page:: 6 hl-color:: green id:: 64d27134-a9da-425a-846e-8b4f0499a9c4
• We leave the according study of performance and scalability of our proposed approach as future work ls-type:: annotation hl-page:: 6 hl-color:: green id:: 64d27142-6f36-4af5-8588-9a1759a20330