Work

Aiおよび量子ソフトウェアシステムに対するデバッギング

Year: 2026
Type: dissertation
Abstract: In recent years, software systems enhanced with the capabilities of artificial intelligence (AI) and quantum computation have attracted increasing attention.We refer to these systems as AI-enabled software systems and quantum software systems, respectively.These systems enable tasks that were difficult or even impossible to accomplish with conventional software.For example, AI-enabled software systems include applications such as autonomous driving and chatbots through inductive development paradigms that learn from large amounts of data.On the other hand, quantum software systems are expected to enhance applications such as optimization and cryptography by leveraging their computational power.Given their increasing deployment in critical domains, misbehavior in these systems can lead to threats to human life or substantial economic losses.This underscores the need for rigorous software engineering practices for these software systems, just as for conventional software systems.Among software engineering practices, debugging, the process of isolating and correcting mistakes in computer software, is essential for ensuring software quality.To support this process, the software engineering research community has dedicated considerable effort, including automated fault localization and automated program repair.However, conventional debugging techniques may not be well suited to emerging software paradigms such as AI-enabled and quantum software systems.For AI-enabled software systems, debugging requires attention not only to the source code but also to the datasets and trained AI models, as their behavior depends on data-driven learning processes.For quantum software systems, debugging should take into account the quantum mechanical properties of quantum software, such as superposition and entanglement of quantum bits (qubits).These challenges motivate the development of novel debugging methodologies tailored to AI-enabled and quantum software systems.This dissertation addresses the challenges of debugging AI-enabled and quantum software systems through four complementary studies related to fault localization and repair.The main contributions of this dissertation are as follows: the first two concern debugging for AI-enabled software systems, and the remaining two concern debugging for quantum software systems.
Institution: Kyushu University
Language: English
Subfield: Software
Open Access status: green
OpenAlex ID: W7225289485
Cites: 188
Cited by:
Related to: 10